From 1ceb01665fff846560fe4679521639985303878b Mon Sep 17 00:00:00 2001
From: Om Chauhan <omchauhan64408@gmail.com>
Date: Sun, 4 Jan 2026 11:04:30 +0530
Subject: [PATCH 01/51] fix: treat language as first-class STT setting

---
 src/pipecat/services/stt_service.py | 2 ++
 1 file changed, 2 insertions(+)

diff --git a/src/pipecat/services/stt_service.py b/src/pipecat/services/stt_service.py
index 6c5741bfe..953ae8dc8 100644
--- a/src/pipecat/services/stt_service.py
+++ b/src/pipecat/services/stt_service.py
@@ -152,6 +152,8 @@ class STTService(AIService):
                 self._settings[key] = value
                 if key == "language":
                     await self.set_language(value)
+            elif key == "language":
+                await self.set_language(value)
             elif key == "model":
                 self.set_model_name(value)
             else:

From 38506f51f7936902d2411530fd4a338dc8d83a43 Mon Sep 17 00:00:00 2001
From: Om Chauhan <omchauhan64408@gmail.com>
Date: Sun, 11 Jan 2026 21:19:47 +0530
Subject: [PATCH 02/51] fix(openrouter): handle multiple system messages for
 Gemini models

---
 src/pipecat/services/openrouter/llm.py | 34 +++++++++++++++++++++++++-
 1 file changed, 33 insertions(+), 1 deletion(-)

diff --git a/src/pipecat/services/openrouter/llm.py b/src/pipecat/services/openrouter/llm.py
index 62992eb23..a86b18573 100644
--- a/src/pipecat/services/openrouter/llm.py
+++ b/src/pipecat/services/openrouter/llm.py
@@ -10,7 +10,7 @@ This module provides an OpenAI-compatible interface for interacting with OpenRou
 extending the base OpenAI LLM service functionality.
 """
 
-from typing import Optional
+from typing import Any, Dict, Optional
 
 from loguru import logger
 
@@ -61,3 +61,35 @@ class OpenRouterLLMService(OpenAILLMService):
         """
         logger.debug(f"Creating OpenRouter client with api {base_url}")
         return super().create_client(api_key, base_url, **kwargs)
+
+    def build_chat_completion_params(self, params_from_context: Dict[str, Any]) -> Dict[str, Any]:
+        """Builds chat parameters, handling model-specific constraints.
+
+        Args:
+            params_from_context: Parameters from the LLM context.
+
+        Returns:
+            Transformed parameters ready for the API call.
+        """
+        params = super().build_chat_completion_params(params_from_context)
+        model = getattr(self, "model_name", getattr(self, "model", "")).lower()
+        if "gemini" in model:
+            messages = params.get("messages", [])
+            if not messages:
+                return params
+            transformed_messages = []
+            system_message_seen = False
+            for msg in messages:
+                if msg.get("role") == "system":
+                    if not system_message_seen:
+                        transformed_messages.append(msg)
+                        system_message_seen = True
+                    else:
+                        new_msg = msg.copy()
+                        new_msg["role"] = "user"
+                        transformed_messages.append(new_msg)
+                else:
+                    transformed_messages.append(msg)
+            params["messages"] = transformed_messages
+
+        return params

From 6fa797c8e48e7efe89f64b27da81cfbbc7860721 Mon Sep 17 00:00:00 2001
From: Paul Kompfner <paul@daily.co>
Date: Fri, 16 Jan 2026 22:01:39 -0500
Subject: [PATCH 03/51] Fix AWS Nova Sonic `reset_conversation()`, which would
 previously error out.

Issues:
- After disconnecting, we were prematurely sending audio messages using the new prompt and content names, before the new prompt and content were created
- We weren't properly sending system instruction and conversation history messages to Nova Sonic with `"interactive": false`
---
 src/pipecat/services/aws/nova_sonic/llm.py | 29 ++++++++++++++++++----
 1 file changed, 24 insertions(+), 5 deletions(-)

diff --git a/src/pipecat/services/aws/nova_sonic/llm.py b/src/pipecat/services/aws/nova_sonic/llm.py
index fbcbe292e..e159ae9f6 100644
--- a/src/pipecat/services/aws/nova_sonic/llm.py
+++ b/src/pipecat/services/aws/nova_sonic/llm.py
@@ -296,6 +296,7 @@ class AWSNovaSonicLLMService(LLMService):
         self._user_text_buffer = ""
         self._assistant_text_buffer = ""
         self._completed_tool_calls = set()
+        self._audio_input_started = False
 
         file_path = files("pipecat.services.aws.nova_sonic").joinpath("ready.wav")
         with wave.open(file_path.open("rb"), "rb") as wav_file:
@@ -533,9 +534,16 @@ class AWSNovaSonicLLMService(LLMService):
             await self._send_text_event(text=system_instruction, role=Role.SYSTEM)
 
         # Send conversation history
-        for message in llm_connection_params["messages"]:
+        messages = llm_connection_params["messages"]
+        for i, message in enumerate(messages):
             # logger.debug(f"Seeding conversation history with message: {message}")
-            await self._send_text_event(text=message.text, role=message.role)
+            # If last message is from user, mark it as interactive to trigger
+            # bot response
+            is_last_message = i == len(messages) - 1
+            interactive = is_last_message and message.role == Role.USER
+            await self._send_text_event(
+                text=message.text, role=message.role, interactive=interactive
+            )
 
         # Start audio input
         await self._send_audio_input_start_event()
@@ -602,6 +610,7 @@ class AWSNovaSonicLLMService(LLMService):
             self._user_text_buffer = ""
             self._assistant_text_buffer = ""
             self._completed_tool_calls = set()
+            self._audio_input_started = False
 
             logger.info("Finished disconnecting")
         except Exception as e:
@@ -727,8 +736,18 @@ class AWSNovaSonicLLMService(LLMService):
         }}
         '''
         await self._send_client_event(audio_content_start)
+        self._audio_input_started = True
 
-    async def _send_text_event(self, text: str, role: Role):
+    async def _send_text_event(self, text: str, role: Role, interactive: bool = False):
+        """Send a text event to the LLM.
+
+        Args:
+            text: The text content to send.
+            role: The role associated with the text (e.g., USER, ASSISTANT, SYSTEM).
+            interactive: Whether the content is interactive. Defaults to False.
+                False: conversation history or system instruction, sent prior to interactive audio
+                True: text input sent during (or at the start of) interactive audio
+        """
         if not self._stream or not self._prompt_name or not text:
             return
 
@@ -741,7 +760,7 @@ class AWSNovaSonicLLMService(LLMService):
                     "promptName": "{self._prompt_name}",
                     "contentName": "{content_name}",
                     "type": "TEXT",
-                    "interactive": true,
+                    "interactive": {json.dumps(interactive)},
                     "role": "{role.value}",
                     "textInputConfiguration": {{
                         "mediaType": "text/plain"
@@ -779,7 +798,7 @@ class AWSNovaSonicLLMService(LLMService):
         await self._send_client_event(text_content_end)
 
     async def _send_user_audio_event(self, audio: bytes):
-        if not self._stream:
+        if not self._stream or not self._audio_input_started:
             return
 
         blob = base64.b64encode(audio)

From 4a9eb82f921ba77bbf2dc533f59552cca34f0f7f Mon Sep 17 00:00:00 2001
From: Om Chauhan <omchauhan64408@gmail.com>
Date: Sun, 18 Jan 2026 20:39:13 +0530
Subject: [PATCH 04/51] fix: preserve UninterruptibleFrames in
 __reset_process_queue

---
 src/pipecat/processors/frame_processor.py | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

diff --git a/src/pipecat/processors/frame_processor.py b/src/pipecat/processors/frame_processor.py
index 9c26fe382..70f44dfca 100644
--- a/src/pipecat/processors/frame_processor.py
+++ b/src/pipecat/processors/frame_processor.py
@@ -950,7 +950,8 @@ class FrameProcessor(BaseObject):
         # Process current queue and keep UninterruptibleFrame frames.
         while not self.__process_queue.empty():
             item = self.__process_queue.get_nowait()
-            if isinstance(item, UninterruptibleFrame):
+            frame = item[0]
+            if isinstance(frame, UninterruptibleFrame):
                 new_queue.put_nowait(item)
             self.__process_queue.task_done()
 

From dfc1f09b772e68a182d4a9a349475538d35a10eb Mon Sep 17 00:00:00 2001
From: Luke Payyapilli <luke.payyapilli@gmail.com>
Date: Mon, 19 Jan 2026 11:00:23 -0500
Subject: [PATCH 05/51] fix(livekit): prevent memory leak when video_in_enabled
 is False

---
 src/pipecat/transports/livekit/transport.py |  13 +-
 tests/test_livekit_transport.py             | 133 ++++++++++++++++++++
 2 files changed, 141 insertions(+), 5 deletions(-)
 create mode 100644 tests/test_livekit_transport.py

diff --git a/src/pipecat/transports/livekit/transport.py b/src/pipecat/transports/livekit/transport.py
index 10a7d8d5d..ab7325c68 100644
--- a/src/pipecat/transports/livekit/transport.py
+++ b/src/pipecat/transports/livekit/transport.py
@@ -539,11 +539,14 @@ class LiveKitTransportClient:
         elif track.kind == rtc.TrackKind.KIND_VIDEO:
             logger.info(f"Video track subscribed: {track.sid} from participant {participant.sid}")
             self._video_tracks[participant.sid] = track
-            video_stream = rtc.VideoStream(track)
-            self._task_manager.create_task(
-                self._process_video_stream(video_stream, participant.sid),
-                f"{self}::_process_video_stream",
-            )
+            # Only process video stream if video input is enabled to prevent
+            # unbounded queue growth when there is no consumer for video frames.
+            if self._params.video_in_enabled:
+                video_stream = rtc.VideoStream(track)
+                self._task_manager.create_task(
+                    self._process_video_stream(video_stream, participant.sid),
+                    f"{self}::_process_video_stream",
+                )
             await self._callbacks.on_video_track_subscribed(participant.sid)
 
     async def _async_on_track_unsubscribed(
diff --git a/tests/test_livekit_transport.py b/tests/test_livekit_transport.py
new file mode 100644
index 000000000..a61daaaa4
--- /dev/null
+++ b/tests/test_livekit_transport.py
@@ -0,0 +1,133 @@
+#
+# Copyright (c) 2024-2026, Daily
+#
+# SPDX-License-Identifier: BSD 2-Clause License
+#
+
+"""Tests for LiveKit transport video stream handling.
+
+Regression tests for issue #3116: Memory leak when video_in_enabled=False
+but video tracks are subscribed.
+"""
+
+import asyncio
+import unittest
+from unittest.mock import AsyncMock, MagicMock
+
+from livekit import rtc
+
+from pipecat.transports.livekit.transport import (
+    LiveKitCallbacks,
+    LiveKitParams,
+    LiveKitTransportClient,
+)
+
+
+class TestLiveKitTransportClient(unittest.IsolatedAsyncioTestCase):
+    """Tests for LiveKitTransportClient video stream handling."""
+
+    def _create_client(self, video_in_enabled: bool) -> LiveKitTransportClient:
+        """Create a LiveKitTransportClient with the specified video_in_enabled setting."""
+        params = LiveKitParams(video_in_enabled=video_in_enabled)
+
+        callbacks = LiveKitCallbacks(
+            on_connected=AsyncMock(),
+            on_disconnected=AsyncMock(),
+            on_before_disconnect=AsyncMock(),
+            on_participant_connected=AsyncMock(),
+            on_participant_disconnected=AsyncMock(),
+            on_audio_track_subscribed=AsyncMock(),
+            on_audio_track_unsubscribed=AsyncMock(),
+            on_video_track_subscribed=AsyncMock(),
+            on_video_track_unsubscribed=AsyncMock(),
+            on_data_received=AsyncMock(),
+            on_first_participant_joined=AsyncMock(),
+        )
+
+        client = LiveKitTransportClient(
+            url="wss://test.livekit.cloud",
+            token="test-token",
+            room_name="test-room",
+            params=params,
+            callbacks=callbacks,
+            transport_name="test-transport",
+        )
+
+        # Mock the task manager
+        client._task_manager = MagicMock()
+        client._task_manager.create_task = MagicMock()
+
+        return client
+
+    def _create_mock_video_track(self) -> tuple:
+        """Create mock video track, publication, and participant."""
+        mock_track = MagicMock()
+        mock_track.kind = rtc.TrackKind.KIND_VIDEO
+        mock_track.sid = "test-track-sid"
+
+        mock_publication = MagicMock()
+
+        mock_participant = MagicMock()
+        mock_participant.sid = "test-participant-sid"
+
+        return mock_track, mock_publication, mock_participant
+
+    def _was_video_stream_task_created(self, client: LiveKitTransportClient) -> bool:
+        """Check if _process_video_stream task was created."""
+        for call in client._task_manager.create_task.call_args_list:
+            task_name = call[0][1] if len(call[0]) > 1 else call[1].get("name", "")
+            if "_process_video_stream" in task_name:
+                return True
+        return False
+
+    async def test_video_stream_not_started_when_video_in_disabled(self):
+        """Test that _process_video_stream is NOT started when video_in_enabled=False.
+
+        This prevents unbounded queue growth when there is no consumer for video frames.
+        Regression test for issue #3116.
+        """
+        client = self._create_client(video_in_enabled=False)
+        mock_track, mock_publication, mock_participant = self._create_mock_video_track()
+
+        # Call the track subscribed handler
+        await client._async_on_track_subscribed(mock_track, mock_publication, mock_participant)
+
+        # Verify that create_task was NOT called for video stream processing
+        self.assertFalse(
+            self._was_video_stream_task_created(client),
+            "Video stream processing should NOT be started when video_in_enabled=False",
+        )
+
+        # Verify that the callback was still called
+        client._callbacks.on_video_track_subscribed.assert_called_once_with(mock_participant.sid)
+
+        # Verify that the track was still added to _video_tracks
+        self.assertIn(mock_participant.sid, client._video_tracks)
+
+    async def test_video_stream_started_when_video_in_enabled(self):
+        """Test that _process_video_stream IS started when video_in_enabled=True."""
+        from unittest.mock import patch
+
+        client = self._create_client(video_in_enabled=True)
+        mock_track, mock_publication, mock_participant = self._create_mock_video_track()
+
+        # Mock rtc.VideoStream to avoid needing a real LiveKit connection
+        with patch("pipecat.transports.livekit.transport.rtc.VideoStream"):
+            # Call the track subscribed handler
+            await client._async_on_track_subscribed(mock_track, mock_publication, mock_participant)
+
+        # Verify that create_task WAS called for video stream processing
+        self.assertTrue(
+            self._was_video_stream_task_created(client),
+            "Video stream processing SHOULD be started when video_in_enabled=True",
+        )
+
+        # Verify that the callback was called
+        client._callbacks.on_video_track_subscribed.assert_called_once_with(mock_participant.sid)
+
+        # Verify that the track was added to _video_tracks
+        self.assertIn(mock_participant.sid, client._video_tracks)
+
+
+if __name__ == "__main__":
+    unittest.main()

From 562bdd3084fe064f74a670910e8f7c4a884f815b Mon Sep 17 00:00:00 2001
From: Luke Payyapilli <luke.payyapilli@gmail.com>
Date: Mon, 19 Jan 2026 11:11:54 -0500
Subject: [PATCH 06/51] test: add livekit to dev deps and improve test clarity

---
 pyproject.toml                  |   1 +
 tests/test_livekit_transport.py | 119 +++++++++++++++-----------------
 uv.lock                         |   8 +--
 3 files changed, 57 insertions(+), 71 deletions(-)

diff --git a/pyproject.toml b/pyproject.toml
index e99ab62bc..f86846de3 100644
--- a/pyproject.toml
+++ b/pyproject.toml
@@ -125,6 +125,7 @@ dev = [
     "build~=1.2.2",
     "coverage~=7.9.1",
     "grpcio-tools~=1.67.1",
+    "livekit~=1.0.13",
     "pip-tools~=7.4.1",
     "pre-commit~=4.2.0",
     "pyright>=1.1.404,<1.2",
diff --git a/tests/test_livekit_transport.py b/tests/test_livekit_transport.py
index a61daaaa4..7962e4e8a 100644
--- a/tests/test_livekit_transport.py
+++ b/tests/test_livekit_transport.py
@@ -7,12 +7,12 @@
 """Tests for LiveKit transport video stream handling.
 
 Regression tests for issue #3116: Memory leak when video_in_enabled=False
-but video tracks are subscribed.
+but video tracks are subscribed. The fix ensures video stream processing
+only starts when there is a consumer for the frames.
 """
 
-import asyncio
 import unittest
-from unittest.mock import AsyncMock, MagicMock
+from unittest.mock import AsyncMock, MagicMock, patch
 
 from livekit import rtc
 
@@ -23,13 +23,19 @@ from pipecat.transports.livekit.transport import (
 )
 
 
-class TestLiveKitTransportClient(unittest.IsolatedAsyncioTestCase):
-    """Tests for LiveKitTransportClient video stream handling."""
+class TestLiveKitVideoStreamMemoryLeak(unittest.IsolatedAsyncioTestCase):
+    """Regression tests for video queue memory leak (#3116).
+
+    The bug: When video_in_enabled=False, subscribing to a video track would
+    start a producer that fills _video_queue, but no consumer would drain it,
+    causing unbounded memory growth (~3GB/min).
+
+    The fix: Only start video stream processing when video_in_enabled=True.
+    """
 
     def _create_client(self, video_in_enabled: bool) -> LiveKitTransportClient:
-        """Create a LiveKitTransportClient with the specified video_in_enabled setting."""
+        """Create a client with the specified video input setting."""
         params = LiveKitParams(video_in_enabled=video_in_enabled)
-
         callbacks = LiveKitCallbacks(
             on_connected=AsyncMock(),
             on_disconnected=AsyncMock(),
@@ -43,7 +49,6 @@ class TestLiveKitTransportClient(unittest.IsolatedAsyncioTestCase):
             on_data_received=AsyncMock(),
             on_first_participant_joined=AsyncMock(),
         )
-
         client = LiveKitTransportClient(
             url="wss://test.livekit.cloud",
             token="test-token",
@@ -52,81 +57,65 @@ class TestLiveKitTransportClient(unittest.IsolatedAsyncioTestCase):
             callbacks=callbacks,
             transport_name="test-transport",
         )
-
-        # Mock the task manager
         client._task_manager = MagicMock()
-        client._task_manager.create_task = MagicMock()
-
         return client
 
-    def _create_mock_video_track(self) -> tuple:
-        """Create mock video track, publication, and participant."""
-        mock_track = MagicMock()
-        mock_track.kind = rtc.TrackKind.KIND_VIDEO
-        mock_track.sid = "test-track-sid"
+    def _create_mock_video_track(self):
+        """Create a mock video track subscription event."""
+        track = MagicMock()
+        track.kind = rtc.TrackKind.KIND_VIDEO
+        track.sid = "video-track-123"
+        publication = MagicMock()
+        participant = MagicMock()
+        participant.sid = "participant-456"
+        return track, publication, participant
 
-        mock_publication = MagicMock()
+    async def test_disabled_video_input_does_not_start_queue_producer(self):
+        """When video input is disabled, no producer should fill the queue.
 
-        mock_participant = MagicMock()
-        mock_participant.sid = "test-participant-sid"
-
-        return mock_track, mock_publication, mock_participant
-
-    def _was_video_stream_task_created(self, client: LiveKitTransportClient) -> bool:
-        """Check if _process_video_stream task was created."""
-        for call in client._task_manager.create_task.call_args_list:
-            task_name = call[0][1] if len(call[0]) > 1 else call[1].get("name", "")
-            if "_process_video_stream" in task_name:
-                return True
-        return False
-
-    async def test_video_stream_not_started_when_video_in_disabled(self):
-        """Test that _process_video_stream is NOT started when video_in_enabled=False.
-
-        This prevents unbounded queue growth when there is no consumer for video frames.
-        Regression test for issue #3116.
+        This prevents the memory leak where frames accumulate with no consumer.
         """
         client = self._create_client(video_in_enabled=False)
-        mock_track, mock_publication, mock_participant = self._create_mock_video_track()
+        track, publication, participant = self._create_mock_video_track()
 
-        # Call the track subscribed handler
-        await client._async_on_track_subscribed(mock_track, mock_publication, mock_participant)
+        await client._async_on_track_subscribed(track, publication, participant)
 
-        # Verify that create_task was NOT called for video stream processing
-        self.assertFalse(
-            self._was_video_stream_task_created(client),
-            "Video stream processing should NOT be started when video_in_enabled=False",
-        )
+        # Verify no video processing task was started
+        task_names = [
+            call[0][1] for call in client._task_manager.create_task.call_args_list
+        ]
+        video_tasks = [name for name in task_names if "video" in name.lower()]
+        self.assertEqual(video_tasks, [], "No video processing task should be started")
 
-        # Verify that the callback was still called
-        client._callbacks.on_video_track_subscribed.assert_called_once_with(mock_participant.sid)
+        # Queue should remain empty
+        self.assertEqual(client._video_queue.qsize(), 0)
 
-        # Verify that the track was still added to _video_tracks
-        self.assertIn(mock_participant.sid, client._video_tracks)
+        # Track metadata should still be recorded
+        self.assertIn(participant.sid, client._video_tracks)
 
-    async def test_video_stream_started_when_video_in_enabled(self):
-        """Test that _process_video_stream IS started when video_in_enabled=True."""
-        from unittest.mock import patch
+        # Callback should still fire for user code
+        client._callbacks.on_video_track_subscribed.assert_called_once()
 
+    async def test_enabled_video_input_starts_queue_producer(self):
+        """When video input is enabled, the producer should start."""
         client = self._create_client(video_in_enabled=True)
-        mock_track, mock_publication, mock_participant = self._create_mock_video_track()
+        track, publication, participant = self._create_mock_video_track()
 
-        # Mock rtc.VideoStream to avoid needing a real LiveKit connection
-        with patch("pipecat.transports.livekit.transport.rtc.VideoStream"):
-            # Call the track subscribed handler
-            await client._async_on_track_subscribed(mock_track, mock_publication, mock_participant)
+        with patch.object(rtc, "VideoStream"):
+            await client._async_on_track_subscribed(track, publication, participant)
 
-        # Verify that create_task WAS called for video stream processing
-        self.assertTrue(
-            self._was_video_stream_task_created(client),
-            "Video stream processing SHOULD be started when video_in_enabled=True",
-        )
+        # Verify video processing task was started
+        task_names = [
+            call[0][1] for call in client._task_manager.create_task.call_args_list
+        ]
+        video_tasks = [name for name in task_names if "video" in name.lower()]
+        self.assertEqual(len(video_tasks), 1, "Video processing task should be started")
 
-        # Verify that the callback was called
-        client._callbacks.on_video_track_subscribed.assert_called_once_with(mock_participant.sid)
+        # Track metadata should be recorded
+        self.assertIn(participant.sid, client._video_tracks)
 
-        # Verify that the track was added to _video_tracks
-        self.assertIn(mock_participant.sid, client._video_tracks)
+        # Callback should fire
+        client._callbacks.on_video_track_subscribed.assert_called_once()
 
 
 if __name__ == "__main__":
diff --git a/uv.lock b/uv.lock
index 45c1f892f..374dd0576 100644
--- a/uv.lock
+++ b/uv.lock
@@ -2013,7 +2013,6 @@ wheels = [
     { url = "https://files.pythonhosted.org/packages/32/6a/33d1702184d94106d3cdd7bfb788e19723206fce152e303473ca3b946c7b/greenlet-3.3.0-cp310-cp310-macosx_11_0_universal2.whl", hash = "sha256:6f8496d434d5cb2dce025773ba5597f71f5410ae499d5dd9533e0653258cdb3d", size = 273658, upload-time = "2025-12-04T14:23:37.494Z" },
     { url = "https://files.pythonhosted.org/packages/d6/b7/2b5805bbf1907c26e434f4e448cd8b696a0b71725204fa21a211ff0c04a7/greenlet-3.3.0-cp310-cp310-manylinux_2_24_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:b96dc7eef78fd404e022e165ec55327f935b9b52ff355b067eb4a0267fc1cffb", size = 574810, upload-time = "2025-12-04T14:50:04.154Z" },
     { url = "https://files.pythonhosted.org/packages/94/38/343242ec12eddf3d8458c73f555c084359883d4ddc674240d9e61ec51fd6/greenlet-3.3.0-cp310-cp310-manylinux_2_24_ppc64le.manylinux_2_28_ppc64le.whl", hash = "sha256:73631cd5cccbcfe63e3f9492aaa664d278fda0ce5c3d43aeda8e77317e38efbd", size = 586248, upload-time = "2025-12-04T14:57:39.35Z" },
-    { url = "https://files.pythonhosted.org/packages/f0/d0/0ae86792fb212e4384041e0ef8e7bc66f59a54912ce407d26a966ed2914d/greenlet-3.3.0-cp310-cp310-manylinux_2_24_s390x.manylinux_2_28_s390x.whl", hash = "sha256:b299a0cb979f5d7197442dccc3aee67fce53500cd88951b7e6c35575701c980b", size = 597403, upload-time = "2025-12-04T15:07:10.831Z" },
     { url = "https://files.pythonhosted.org/packages/b6/a8/15d0aa26c0036a15d2659175af00954aaaa5d0d66ba538345bd88013b4d7/greenlet-3.3.0-cp310-cp310-manylinux_2_24_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:7dee147740789a4632cace364816046e43310b59ff8fb79833ab043aefa72fd5", size = 586910, upload-time = "2025-12-04T14:25:59.705Z" },
     { url = "https://files.pythonhosted.org/packages/e1/9b/68d5e3b7ccaba3907e5532cf8b9bf16f9ef5056a008f195a367db0ff32db/greenlet-3.3.0-cp310-cp310-musllinux_1_2_aarch64.whl", hash = "sha256:39b28e339fc3c348427560494e28d8a6f3561c8d2bcf7d706e1c624ed8d822b9", size = 1547206, upload-time = "2025-12-04T15:04:21.027Z" },
     { url = "https://files.pythonhosted.org/packages/66/bd/e3086ccedc61e49f91e2cfb5ffad9d8d62e5dc85e512a6200f096875b60c/greenlet-3.3.0-cp310-cp310-musllinux_1_2_x86_64.whl", hash = "sha256:b3c374782c2935cc63b2a27ba8708471de4ad1abaa862ffdb1ef45a643ddbb7d", size = 1613359, upload-time = "2025-12-04T14:27:26.548Z" },
@@ -2021,7 +2020,6 @@ wheels = [
     { url = "https://files.pythonhosted.org/packages/1f/cb/48e964c452ca2b92175a9b2dca037a553036cb053ba69e284650ce755f13/greenlet-3.3.0-cp311-cp311-macosx_11_0_universal2.whl", hash = "sha256:e29f3018580e8412d6aaf5641bb7745d38c85228dacf51a73bd4e26ddf2a6a8e", size = 274908, upload-time = "2025-12-04T14:23:26.435Z" },
     { url = "https://files.pythonhosted.org/packages/28/da/38d7bff4d0277b594ec557f479d65272a893f1f2a716cad91efeb8680953/greenlet-3.3.0-cp311-cp311-manylinux_2_24_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:a687205fb22794e838f947e2194c0566d3812966b41c78709554aa883183fb62", size = 577113, upload-time = "2025-12-04T14:50:05.493Z" },
     { url = "https://files.pythonhosted.org/packages/3c/f2/89c5eb0faddc3ff014f1c04467d67dee0d1d334ab81fadbf3744847f8a8a/greenlet-3.3.0-cp311-cp311-manylinux_2_24_ppc64le.manylinux_2_28_ppc64le.whl", hash = "sha256:4243050a88ba61842186cb9e63c7dfa677ec146160b0efd73b855a3d9c7fcf32", size = 590338, upload-time = "2025-12-04T14:57:41.136Z" },
-    { url = "https://files.pythonhosted.org/packages/80/d7/db0a5085035d05134f8c089643da2b44cc9b80647c39e93129c5ef170d8f/greenlet-3.3.0-cp311-cp311-manylinux_2_24_s390x.manylinux_2_28_s390x.whl", hash = "sha256:670d0f94cd302d81796e37299bcd04b95d62403883b24225c6b5271466612f45", size = 601098, upload-time = "2025-12-04T15:07:11.898Z" },
     { url = "https://files.pythonhosted.org/packages/dc/a6/e959a127b630a58e23529972dbc868c107f9d583b5a9f878fb858c46bc1a/greenlet-3.3.0-cp311-cp311-manylinux_2_24_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:6cb3a8ec3db4a3b0eb8a3c25436c2d49e3505821802074969db017b87bc6a948", size = 590206, upload-time = "2025-12-04T14:26:01.254Z" },
     { url = "https://files.pythonhosted.org/packages/48/60/29035719feb91798693023608447283b266b12efc576ed013dd9442364bb/greenlet-3.3.0-cp311-cp311-musllinux_1_2_aarch64.whl", hash = "sha256:2de5a0b09eab81fc6a382791b995b1ccf2b172a9fec934747a7a23d2ff291794", size = 1550668, upload-time = "2025-12-04T15:04:22.439Z" },
     { url = "https://files.pythonhosted.org/packages/0a/5f/783a23754b691bfa86bd72c3033aa107490deac9b2ef190837b860996c9f/greenlet-3.3.0-cp311-cp311-musllinux_1_2_x86_64.whl", hash = "sha256:4449a736606bd30f27f8e1ff4678ee193bc47f6ca810d705981cfffd6ce0d8c5", size = 1615483, upload-time = "2025-12-04T14:27:28.083Z" },
@@ -2029,7 +2027,6 @@ wheels = [
     { url = "https://files.pythonhosted.org/packages/f8/0a/a3871375c7b9727edaeeea994bfff7c63ff7804c9829c19309ba2e058807/greenlet-3.3.0-cp312-cp312-macosx_11_0_universal2.whl", hash = "sha256:b01548f6e0b9e9784a2c99c5651e5dc89ffcbe870bc5fb2e5ef864e9cc6b5dcb", size = 276379, upload-time = "2025-12-04T14:23:30.498Z" },
     { url = "https://files.pythonhosted.org/packages/43/ab/7ebfe34dce8b87be0d11dae91acbf76f7b8246bf9d6b319c741f99fa59c6/greenlet-3.3.0-cp312-cp312-manylinux_2_24_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:349345b770dc88f81506c6861d22a6ccd422207829d2c854ae2af8025af303e3", size = 597294, upload-time = "2025-12-04T14:50:06.847Z" },
     { url = "https://files.pythonhosted.org/packages/a4/39/f1c8da50024feecd0793dbd5e08f526809b8ab5609224a2da40aad3a7641/greenlet-3.3.0-cp312-cp312-manylinux_2_24_ppc64le.manylinux_2_28_ppc64le.whl", hash = "sha256:e8e18ed6995e9e2c0b4ed264d2cf89260ab3ac7e13555b8032b25a74c6d18655", size = 607742, upload-time = "2025-12-04T14:57:42.349Z" },
-    { url = "https://files.pythonhosted.org/packages/77/cb/43692bcd5f7a0da6ec0ec6d58ee7cddb606d055ce94a62ac9b1aa481e969/greenlet-3.3.0-cp312-cp312-manylinux_2_24_s390x.manylinux_2_28_s390x.whl", hash = "sha256:c024b1e5696626890038e34f76140ed1daf858e37496d33f2af57f06189e70d7", size = 622297, upload-time = "2025-12-04T15:07:13.552Z" },
     { url = "https://files.pythonhosted.org/packages/75/b0/6bde0b1011a60782108c01de5913c588cf51a839174538d266de15e4bf4d/greenlet-3.3.0-cp312-cp312-manylinux_2_24_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:047ab3df20ede6a57c35c14bf5200fcf04039d50f908270d3f9a7a82064f543b", size = 609885, upload-time = "2025-12-04T14:26:02.368Z" },
     { url = "https://files.pythonhosted.org/packages/49/0e/49b46ac39f931f59f987b7cd9f34bfec8ef81d2a1e6e00682f55be5de9f4/greenlet-3.3.0-cp312-cp312-musllinux_1_2_aarch64.whl", hash = "sha256:2d9ad37fc657b1102ec880e637cccf20191581f75c64087a549e66c57e1ceb53", size = 1567424, upload-time = "2025-12-04T15:04:23.757Z" },
     { url = "https://files.pythonhosted.org/packages/05/f5/49a9ac2dff7f10091935def9165c90236d8f175afb27cbed38fb1d61ab6b/greenlet-3.3.0-cp312-cp312-musllinux_1_2_x86_64.whl", hash = "sha256:83cd0e36932e0e7f36a64b732a6f60c2fc2df28c351bae79fbaf4f8092fe7614", size = 1636017, upload-time = "2025-12-04T14:27:29.688Z" },
@@ -2037,7 +2034,6 @@ wheels = [
     { url = "https://files.pythonhosted.org/packages/02/2f/28592176381b9ab2cafa12829ba7b472d177f3acc35d8fbcf3673d966fff/greenlet-3.3.0-cp313-cp313-macosx_11_0_universal2.whl", hash = "sha256:a1e41a81c7e2825822f4e068c48cb2196002362619e2d70b148f20a831c00739", size = 275140, upload-time = "2025-12-04T14:23:01.282Z" },
     { url = "https://files.pythonhosted.org/packages/2c/80/fbe937bf81e9fca98c981fe499e59a3f45df2a04da0baa5c2be0dca0d329/greenlet-3.3.0-cp313-cp313-manylinux_2_24_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:9f515a47d02da4d30caaa85b69474cec77b7929b2e936ff7fb853d42f4bf8808", size = 599219, upload-time = "2025-12-04T14:50:08.309Z" },
     { url = "https://files.pythonhosted.org/packages/c2/ff/7c985128f0514271b8268476af89aee6866df5eec04ac17dcfbc676213df/greenlet-3.3.0-cp313-cp313-manylinux_2_24_ppc64le.manylinux_2_28_ppc64le.whl", hash = "sha256:7d2d9fd66bfadf230b385fdc90426fcd6eb64db54b40c495b72ac0feb5766c54", size = 610211, upload-time = "2025-12-04T14:57:43.968Z" },
-    { url = "https://files.pythonhosted.org/packages/79/07/c47a82d881319ec18a4510bb30463ed6891f2ad2c1901ed5ec23d3de351f/greenlet-3.3.0-cp313-cp313-manylinux_2_24_s390x.manylinux_2_28_s390x.whl", hash = "sha256:30a6e28487a790417d036088b3bcb3f3ac7d8babaa7d0139edbaddebf3af9492", size = 624311, upload-time = "2025-12-04T15:07:14.697Z" },
     { url = "https://files.pythonhosted.org/packages/fd/8e/424b8c6e78bd9837d14ff7df01a9829fc883ba2ab4ea787d4f848435f23f/greenlet-3.3.0-cp313-cp313-manylinux_2_24_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:087ea5e004437321508a8d6f20efc4cfec5e3c30118e1417ea96ed1d93950527", size = 612833, upload-time = "2025-12-04T14:26:03.669Z" },
     { url = "https://files.pythonhosted.org/packages/b5/ba/56699ff9b7c76ca12f1cdc27a886d0f81f2189c3455ff9f65246780f713d/greenlet-3.3.0-cp313-cp313-musllinux_1_2_aarch64.whl", hash = "sha256:ab97cf74045343f6c60a39913fa59710e4bd26a536ce7ab2397adf8b27e67c39", size = 1567256, upload-time = "2025-12-04T15:04:25.276Z" },
     { url = "https://files.pythonhosted.org/packages/1e/37/f31136132967982d698c71a281a8901daf1a8fbab935dce7c0cf15f942cc/greenlet-3.3.0-cp313-cp313-musllinux_1_2_x86_64.whl", hash = "sha256:5375d2e23184629112ca1ea89a53389dddbffcf417dad40125713d88eb5f96e8", size = 1636483, upload-time = "2025-12-04T14:27:30.804Z" },
@@ -2045,7 +2041,6 @@ wheels = [
     { url = "https://files.pythonhosted.org/packages/d7/7c/f0a6d0ede2c7bf092d00bc83ad5bafb7e6ec9b4aab2fbdfa6f134dc73327/greenlet-3.3.0-cp314-cp314-macosx_11_0_universal2.whl", hash = "sha256:60c2ef0f578afb3c8d92ea07ad327f9a062547137afe91f38408f08aacab667f", size = 275671, upload-time = "2025-12-04T14:23:05.267Z" },
     { url = "https://files.pythonhosted.org/packages/44/06/dac639ae1a50f5969d82d2e3dd9767d30d6dbdbab0e1a54010c8fe90263c/greenlet-3.3.0-cp314-cp314-manylinux_2_24_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:0a5d554d0712ba1de0a6c94c640f7aeba3f85b3a6e1f2899c11c2c0428da9365", size = 646360, upload-time = "2025-12-04T14:50:10.026Z" },
     { url = "https://files.pythonhosted.org/packages/e0/94/0fb76fe6c5369fba9bf98529ada6f4c3a1adf19e406a47332245ef0eb357/greenlet-3.3.0-cp314-cp314-manylinux_2_24_ppc64le.manylinux_2_28_ppc64le.whl", hash = "sha256:3a898b1e9c5f7307ebbde4102908e6cbfcb9ea16284a3abe15cab996bee8b9b3", size = 658160, upload-time = "2025-12-04T14:57:45.41Z" },
-    { url = "https://files.pythonhosted.org/packages/93/79/d2c70cae6e823fac36c3bbc9077962105052b7ef81db2f01ec3b9bf17e2b/greenlet-3.3.0-cp314-cp314-manylinux_2_24_s390x.manylinux_2_28_s390x.whl", hash = "sha256:dcd2bdbd444ff340e8d6bdf54d2f206ccddbb3ccfdcd3c25bf4afaa7b8f0cf45", size = 671388, upload-time = "2025-12-04T15:07:15.789Z" },
     { url = "https://files.pythonhosted.org/packages/b8/14/bab308fc2c1b5228c3224ec2bf928ce2e4d21d8046c161e44a2012b5203e/greenlet-3.3.0-cp314-cp314-manylinux_2_24_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:5773edda4dc00e173820722711d043799d3adb4f01731f40619e07ea2750b955", size = 660166, upload-time = "2025-12-04T14:26:05.099Z" },
     { url = "https://files.pythonhosted.org/packages/4b/d2/91465d39164eaa0085177f61983d80ffe746c5a1860f009811d498e7259c/greenlet-3.3.0-cp314-cp314-musllinux_1_2_aarch64.whl", hash = "sha256:ac0549373982b36d5fd5d30beb8a7a33ee541ff98d2b502714a09f1169f31b55", size = 1615193, upload-time = "2025-12-04T15:04:27.041Z" },
     { url = "https://files.pythonhosted.org/packages/42/1b/83d110a37044b92423084d52d5d5a3b3a73cafb51b547e6d7366ff62eff1/greenlet-3.3.0-cp314-cp314-musllinux_1_2_x86_64.whl", hash = "sha256:d198d2d977460358c3b3a4dc844f875d1adb33817f0613f663a656f463764ccc", size = 1683653, upload-time = "2025-12-04T14:27:32.366Z" },
@@ -2053,7 +2048,6 @@ wheels = [
     { url = "https://files.pythonhosted.org/packages/a0/66/bd6317bc5932accf351fc19f177ffba53712a202f9df10587da8df257c7e/greenlet-3.3.0-cp314-cp314t-macosx_11_0_universal2.whl", hash = "sha256:d6ed6f85fae6cdfdb9ce04c9bf7a08d666cfcfb914e7d006f44f840b46741931", size = 282638, upload-time = "2025-12-04T14:25:20.941Z" },
     { url = "https://files.pythonhosted.org/packages/30/cf/cc81cb030b40e738d6e69502ccbd0dd1bced0588e958f9e757945de24404/greenlet-3.3.0-cp314-cp314t-manylinux_2_24_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:d9125050fcf24554e69c4cacb086b87b3b55dc395a8b3ebe6487b045b2614388", size = 651145, upload-time = "2025-12-04T14:50:11.039Z" },
     { url = "https://files.pythonhosted.org/packages/9c/ea/1020037b5ecfe95ca7df8d8549959baceb8186031da83d5ecceff8b08cd2/greenlet-3.3.0-cp314-cp314t-manylinux_2_24_ppc64le.manylinux_2_28_ppc64le.whl", hash = "sha256:87e63ccfa13c0a0f6234ed0add552af24cc67dd886731f2261e46e241608bee3", size = 654236, upload-time = "2025-12-04T14:57:47.007Z" },
-    { url = "https://files.pythonhosted.org/packages/69/cc/1e4bae2e45ca2fa55299f4e85854606a78ecc37fead20d69322f96000504/greenlet-3.3.0-cp314-cp314t-manylinux_2_24_s390x.manylinux_2_28_s390x.whl", hash = "sha256:2662433acbca297c9153a4023fe2161c8dcfdcc91f10433171cf7e7d94ba2221", size = 662506, upload-time = "2025-12-04T15:07:16.906Z" },
     { url = "https://files.pythonhosted.org/packages/57/b9/f8025d71a6085c441a7eaff0fd928bbb275a6633773667023d19179fe815/greenlet-3.3.0-cp314-cp314t-manylinux_2_24_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:3c6e9b9c1527a78520357de498b0e709fb9e2f49c3a513afd5a249007261911b", size = 653783, upload-time = "2025-12-04T14:26:06.225Z" },
     { url = "https://files.pythonhosted.org/packages/f6/c7/876a8c7a7485d5d6b5c6821201d542ef28be645aa024cfe1145b35c120c1/greenlet-3.3.0-cp314-cp314t-musllinux_1_2_aarch64.whl", hash = "sha256:286d093f95ec98fdd92fcb955003b8a3d054b4e2cab3e2707a5039e7b50520fd", size = 1614857, upload-time = "2025-12-04T15:04:28.484Z" },
     { url = "https://files.pythonhosted.org/packages/4f/dc/041be1dff9f23dac5f48a43323cd0789cb798342011c19a248d9c9335536/greenlet-3.3.0-cp314-cp314t-musllinux_1_2_x86_64.whl", hash = "sha256:6c10513330af5b8ae16f023e8ddbfb486ab355d04467c4679c5cfe4659975dd9", size = 1676034, upload-time = "2025-12-04T14:27:33.531Z" },
@@ -4468,6 +4462,7 @@ dev = [
     { name = "build" },
     { name = "coverage" },
     { name = "grpcio-tools" },
+    { name = "livekit" },
     { name = "pip-tools" },
     { name = "pre-commit" },
     { name = "pyright" },
@@ -4606,6 +4601,7 @@ dev = [
     { name = "build", specifier = "~=1.2.2" },
     { name = "coverage", specifier = "~=7.9.1" },
     { name = "grpcio-tools", specifier = "~=1.67.1" },
+    { name = "livekit", specifier = "~=1.0.13" },
     { name = "pip-tools", specifier = "~=7.4.1" },
     { name = "pre-commit", specifier = "~=4.2.0" },
     { name = "pyright", specifier = ">=1.1.404,<1.2" },

From c89ae717feb606a95126ad25d5a8f70484242507 Mon Sep 17 00:00:00 2001
From: Luke Payyapilli <luke.payyapilli@gmail.com>
Date: Mon, 19 Jan 2026 11:13:41 -0500
Subject: [PATCH 07/51] style: fix ruff formatting

---
 tests/test_livekit_transport.py | 8 ++------
 1 file changed, 2 insertions(+), 6 deletions(-)

diff --git a/tests/test_livekit_transport.py b/tests/test_livekit_transport.py
index 7962e4e8a..1a6184d85 100644
--- a/tests/test_livekit_transport.py
+++ b/tests/test_livekit_transport.py
@@ -81,9 +81,7 @@ class TestLiveKitVideoStreamMemoryLeak(unittest.IsolatedAsyncioTestCase):
         await client._async_on_track_subscribed(track, publication, participant)
 
         # Verify no video processing task was started
-        task_names = [
-            call[0][1] for call in client._task_manager.create_task.call_args_list
-        ]
+        task_names = [call[0][1] for call in client._task_manager.create_task.call_args_list]
         video_tasks = [name for name in task_names if "video" in name.lower()]
         self.assertEqual(video_tasks, [], "No video processing task should be started")
 
@@ -105,9 +103,7 @@ class TestLiveKitVideoStreamMemoryLeak(unittest.IsolatedAsyncioTestCase):
             await client._async_on_track_subscribed(track, publication, participant)
 
         # Verify video processing task was started
-        task_names = [
-            call[0][1] for call in client._task_manager.create_task.call_args_list
-        ]
+        task_names = [call[0][1] for call in client._task_manager.create_task.call_args_list]
         video_tasks = [name for name in task_names if "video" in name.lower()]
         self.assertEqual(len(video_tasks), 1, "Video processing task should be started")
 

From aed44c863af04fe8eec7e9272e10a5c703c1e41e Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Aleix=20Conchillo=20Flaqu=C3=A9?= <aleix@daily.co>
Date: Mon, 19 Jan 2026 14:37:00 -0800
Subject: [PATCH 08/51] scripts(eval): give examples to numerical word answers

Some models need extra help.
---
 scripts/evals/eval.py | 7 ++++---
 1 file changed, 4 insertions(+), 3 deletions(-)

diff --git a/scripts/evals/eval.py b/scripts/evals/eval.py
index 7c6a72604..16598b4d7 100644
--- a/scripts/evals/eval.py
+++ b/scripts/evals/eval.py
@@ -293,12 +293,13 @@ async def run_eval_pipeline(
         "You should only call the eval function if:\n"
         "- The user explicitly attempts to answer the question, AND\n"
         f"- Their answer can be cleanly evaluated using: {eval_config.eval}\n"
-        "Ignore greetings, comments, non-answers, or requests for clarification."
+        "Ignore greetings, comments, non-answers, or requests for clarification.\n"
+        "Numerical word answers are allowed (e.g., 'five' is the same as '5').\n"
     )
     if eval_config.eval_speaks_first:
-        system_prompt = f"You are an evaluation agent, be extremly brief. Numerical word answers are allowed. You will start the conversation by saying: '{example_prompt}'. {common_system_prompt}"
+        system_prompt = f"You are an evaluation agent, be extremly brief. You will start the conversation by saying: '{example_prompt}'. {common_system_prompt}"
     else:
-        system_prompt = f"You are an evaluation agent, be extremly brief. Numerical word answers are allowed. First, ask one question: {example_prompt}. {common_system_prompt}"
+        system_prompt = f"You are an evaluation agent, be extremly brief. First, ask one question: {example_prompt}. {common_system_prompt}"
 
     messages = [
         {

From f6359d460eb058a141e6f6fe2c1aa164932b4861 Mon Sep 17 00:00:00 2001
From: Luke Payyapilli <luke.payyapilli@gmail.com>
Date: Tue, 20 Jan 2026 09:16:16 -0500
Subject: [PATCH 09/51] chore: install livekit as optional extra in CI instead
 of dev dep

---
 .github/workflows/coverage.yaml |  2 +-
 .github/workflows/tests.yaml    |  2 +-
 pyproject.toml                  |  1 -
 tests/test_livekit_transport.py | 18 ++++++++++++------
 uv.lock                         |  2 --
 5 files changed, 14 insertions(+), 11 deletions(-)

diff --git a/.github/workflows/coverage.yaml b/.github/workflows/coverage.yaml
index 0dd30f9e5..faca8c03f 100644
--- a/.github/workflows/coverage.yaml
+++ b/.github/workflows/coverage.yaml
@@ -33,7 +33,7 @@ jobs:
 
       - name: Install dependencies
         run: |
-          uv sync --group dev --extra anthropic --extra aws --extra google --extra langchain --extra websocket
+          uv sync --group dev --extra anthropic --extra aws --extra google --extra langchain --extra livekit --extra websocket
 
       - name: Run tests with coverage
         run: |
diff --git a/.github/workflows/tests.yaml b/.github/workflows/tests.yaml
index 8e58845e4..459725da7 100644
--- a/.github/workflows/tests.yaml
+++ b/.github/workflows/tests.yaml
@@ -37,7 +37,7 @@ jobs:
 
       - name: Install dependencies
         run: |
-          uv sync --group dev --extra anthropic --extra aws --extra google --extra langchain --extra websocket
+          uv sync --group dev --extra anthropic --extra aws --extra google --extra langchain --extra livekit --extra websocket
 
       - name: Test with pytest
         run: |
diff --git a/pyproject.toml b/pyproject.toml
index f86846de3..e99ab62bc 100644
--- a/pyproject.toml
+++ b/pyproject.toml
@@ -125,7 +125,6 @@ dev = [
     "build~=1.2.2",
     "coverage~=7.9.1",
     "grpcio-tools~=1.67.1",
-    "livekit~=1.0.13",
     "pip-tools~=7.4.1",
     "pre-commit~=4.2.0",
     "pyright>=1.1.404,<1.2",
diff --git a/tests/test_livekit_transport.py b/tests/test_livekit_transport.py
index 1a6184d85..472d60cba 100644
--- a/tests/test_livekit_transport.py
+++ b/tests/test_livekit_transport.py
@@ -14,15 +14,21 @@ only starts when there is a consumer for the frames.
 import unittest
 from unittest.mock import AsyncMock, MagicMock, patch
 
-from livekit import rtc
+try:
+    from livekit import rtc
 
-from pipecat.transports.livekit.transport import (
-    LiveKitCallbacks,
-    LiveKitParams,
-    LiveKitTransportClient,
-)
+    from pipecat.transports.livekit.transport import (
+        LiveKitCallbacks,
+        LiveKitParams,
+        LiveKitTransportClient,
+    )
+
+    LIVEKIT_AVAILABLE = True
+except ImportError:
+    LIVEKIT_AVAILABLE = False
 
 
+@unittest.skipUnless(LIVEKIT_AVAILABLE, "livekit package not installed")
 class TestLiveKitVideoStreamMemoryLeak(unittest.IsolatedAsyncioTestCase):
     """Regression tests for video queue memory leak (#3116).
 
diff --git a/uv.lock b/uv.lock
index 374dd0576..2610e2df5 100644
--- a/uv.lock
+++ b/uv.lock
@@ -4462,7 +4462,6 @@ dev = [
     { name = "build" },
     { name = "coverage" },
     { name = "grpcio-tools" },
-    { name = "livekit" },
     { name = "pip-tools" },
     { name = "pre-commit" },
     { name = "pyright" },
@@ -4601,7 +4600,6 @@ dev = [
     { name = "build", specifier = "~=1.2.2" },
     { name = "coverage", specifier = "~=7.9.1" },
     { name = "grpcio-tools", specifier = "~=1.67.1" },
-    { name = "livekit", specifier = "~=1.0.13" },
     { name = "pip-tools", specifier = "~=7.4.1" },
     { name = "pre-commit", specifier = "~=4.2.0" },
     { name = "pyright", specifier = ">=1.1.404,<1.2" },

From 1ac811ab32c46dcc9505361734cc4e74482e8b74 Mon Sep 17 00:00:00 2001
From: Luke Payyapilli <luke.payyapilli@gmail.com>
Date: Tue, 20 Jan 2026 09:19:43 -0500
Subject: [PATCH 10/51] chore: revert unrelated uv.lock changes

---
 uv.lock | 6 ++++++
 1 file changed, 6 insertions(+)

diff --git a/uv.lock b/uv.lock
index 2610e2df5..45c1f892f 100644
--- a/uv.lock
+++ b/uv.lock
@@ -2013,6 +2013,7 @@ wheels = [
     { url = "https://files.pythonhosted.org/packages/32/6a/33d1702184d94106d3cdd7bfb788e19723206fce152e303473ca3b946c7b/greenlet-3.3.0-cp310-cp310-macosx_11_0_universal2.whl", hash = "sha256:6f8496d434d5cb2dce025773ba5597f71f5410ae499d5dd9533e0653258cdb3d", size = 273658, upload-time = "2025-12-04T14:23:37.494Z" },
     { url = "https://files.pythonhosted.org/packages/d6/b7/2b5805bbf1907c26e434f4e448cd8b696a0b71725204fa21a211ff0c04a7/greenlet-3.3.0-cp310-cp310-manylinux_2_24_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:b96dc7eef78fd404e022e165ec55327f935b9b52ff355b067eb4a0267fc1cffb", size = 574810, upload-time = "2025-12-04T14:50:04.154Z" },
     { url = "https://files.pythonhosted.org/packages/94/38/343242ec12eddf3d8458c73f555c084359883d4ddc674240d9e61ec51fd6/greenlet-3.3.0-cp310-cp310-manylinux_2_24_ppc64le.manylinux_2_28_ppc64le.whl", hash = "sha256:73631cd5cccbcfe63e3f9492aaa664d278fda0ce5c3d43aeda8e77317e38efbd", size = 586248, upload-time = "2025-12-04T14:57:39.35Z" },
+    { url = "https://files.pythonhosted.org/packages/f0/d0/0ae86792fb212e4384041e0ef8e7bc66f59a54912ce407d26a966ed2914d/greenlet-3.3.0-cp310-cp310-manylinux_2_24_s390x.manylinux_2_28_s390x.whl", hash = "sha256:b299a0cb979f5d7197442dccc3aee67fce53500cd88951b7e6c35575701c980b", size = 597403, upload-time = "2025-12-04T15:07:10.831Z" },
     { url = "https://files.pythonhosted.org/packages/b6/a8/15d0aa26c0036a15d2659175af00954aaaa5d0d66ba538345bd88013b4d7/greenlet-3.3.0-cp310-cp310-manylinux_2_24_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:7dee147740789a4632cace364816046e43310b59ff8fb79833ab043aefa72fd5", size = 586910, upload-time = "2025-12-04T14:25:59.705Z" },
     { url = "https://files.pythonhosted.org/packages/e1/9b/68d5e3b7ccaba3907e5532cf8b9bf16f9ef5056a008f195a367db0ff32db/greenlet-3.3.0-cp310-cp310-musllinux_1_2_aarch64.whl", hash = "sha256:39b28e339fc3c348427560494e28d8a6f3561c8d2bcf7d706e1c624ed8d822b9", size = 1547206, upload-time = "2025-12-04T15:04:21.027Z" },
     { url = "https://files.pythonhosted.org/packages/66/bd/e3086ccedc61e49f91e2cfb5ffad9d8d62e5dc85e512a6200f096875b60c/greenlet-3.3.0-cp310-cp310-musllinux_1_2_x86_64.whl", hash = "sha256:b3c374782c2935cc63b2a27ba8708471de4ad1abaa862ffdb1ef45a643ddbb7d", size = 1613359, upload-time = "2025-12-04T14:27:26.548Z" },
@@ -2020,6 +2021,7 @@ wheels = [
     { url = "https://files.pythonhosted.org/packages/1f/cb/48e964c452ca2b92175a9b2dca037a553036cb053ba69e284650ce755f13/greenlet-3.3.0-cp311-cp311-macosx_11_0_universal2.whl", hash = "sha256:e29f3018580e8412d6aaf5641bb7745d38c85228dacf51a73bd4e26ddf2a6a8e", size = 274908, upload-time = "2025-12-04T14:23:26.435Z" },
     { url = "https://files.pythonhosted.org/packages/28/da/38d7bff4d0277b594ec557f479d65272a893f1f2a716cad91efeb8680953/greenlet-3.3.0-cp311-cp311-manylinux_2_24_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:a687205fb22794e838f947e2194c0566d3812966b41c78709554aa883183fb62", size = 577113, upload-time = "2025-12-04T14:50:05.493Z" },
     { url = "https://files.pythonhosted.org/packages/3c/f2/89c5eb0faddc3ff014f1c04467d67dee0d1d334ab81fadbf3744847f8a8a/greenlet-3.3.0-cp311-cp311-manylinux_2_24_ppc64le.manylinux_2_28_ppc64le.whl", hash = "sha256:4243050a88ba61842186cb9e63c7dfa677ec146160b0efd73b855a3d9c7fcf32", size = 590338, upload-time = "2025-12-04T14:57:41.136Z" },
+    { url = "https://files.pythonhosted.org/packages/80/d7/db0a5085035d05134f8c089643da2b44cc9b80647c39e93129c5ef170d8f/greenlet-3.3.0-cp311-cp311-manylinux_2_24_s390x.manylinux_2_28_s390x.whl", hash = "sha256:670d0f94cd302d81796e37299bcd04b95d62403883b24225c6b5271466612f45", size = 601098, upload-time = "2025-12-04T15:07:11.898Z" },
     { url = "https://files.pythonhosted.org/packages/dc/a6/e959a127b630a58e23529972dbc868c107f9d583b5a9f878fb858c46bc1a/greenlet-3.3.0-cp311-cp311-manylinux_2_24_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:6cb3a8ec3db4a3b0eb8a3c25436c2d49e3505821802074969db017b87bc6a948", size = 590206, upload-time = "2025-12-04T14:26:01.254Z" },
     { url = "https://files.pythonhosted.org/packages/48/60/29035719feb91798693023608447283b266b12efc576ed013dd9442364bb/greenlet-3.3.0-cp311-cp311-musllinux_1_2_aarch64.whl", hash = "sha256:2de5a0b09eab81fc6a382791b995b1ccf2b172a9fec934747a7a23d2ff291794", size = 1550668, upload-time = "2025-12-04T15:04:22.439Z" },
     { url = "https://files.pythonhosted.org/packages/0a/5f/783a23754b691bfa86bd72c3033aa107490deac9b2ef190837b860996c9f/greenlet-3.3.0-cp311-cp311-musllinux_1_2_x86_64.whl", hash = "sha256:4449a736606bd30f27f8e1ff4678ee193bc47f6ca810d705981cfffd6ce0d8c5", size = 1615483, upload-time = "2025-12-04T14:27:28.083Z" },
@@ -2027,6 +2029,7 @@ wheels = [
     { url = "https://files.pythonhosted.org/packages/f8/0a/a3871375c7b9727edaeeea994bfff7c63ff7804c9829c19309ba2e058807/greenlet-3.3.0-cp312-cp312-macosx_11_0_universal2.whl", hash = "sha256:b01548f6e0b9e9784a2c99c5651e5dc89ffcbe870bc5fb2e5ef864e9cc6b5dcb", size = 276379, upload-time = "2025-12-04T14:23:30.498Z" },
     { url = "https://files.pythonhosted.org/packages/43/ab/7ebfe34dce8b87be0d11dae91acbf76f7b8246bf9d6b319c741f99fa59c6/greenlet-3.3.0-cp312-cp312-manylinux_2_24_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:349345b770dc88f81506c6861d22a6ccd422207829d2c854ae2af8025af303e3", size = 597294, upload-time = "2025-12-04T14:50:06.847Z" },
     { url = "https://files.pythonhosted.org/packages/a4/39/f1c8da50024feecd0793dbd5e08f526809b8ab5609224a2da40aad3a7641/greenlet-3.3.0-cp312-cp312-manylinux_2_24_ppc64le.manylinux_2_28_ppc64le.whl", hash = "sha256:e8e18ed6995e9e2c0b4ed264d2cf89260ab3ac7e13555b8032b25a74c6d18655", size = 607742, upload-time = "2025-12-04T14:57:42.349Z" },
+    { url = "https://files.pythonhosted.org/packages/77/cb/43692bcd5f7a0da6ec0ec6d58ee7cddb606d055ce94a62ac9b1aa481e969/greenlet-3.3.0-cp312-cp312-manylinux_2_24_s390x.manylinux_2_28_s390x.whl", hash = "sha256:c024b1e5696626890038e34f76140ed1daf858e37496d33f2af57f06189e70d7", size = 622297, upload-time = "2025-12-04T15:07:13.552Z" },
     { url = "https://files.pythonhosted.org/packages/75/b0/6bde0b1011a60782108c01de5913c588cf51a839174538d266de15e4bf4d/greenlet-3.3.0-cp312-cp312-manylinux_2_24_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:047ab3df20ede6a57c35c14bf5200fcf04039d50f908270d3f9a7a82064f543b", size = 609885, upload-time = "2025-12-04T14:26:02.368Z" },
     { url = "https://files.pythonhosted.org/packages/49/0e/49b46ac39f931f59f987b7cd9f34bfec8ef81d2a1e6e00682f55be5de9f4/greenlet-3.3.0-cp312-cp312-musllinux_1_2_aarch64.whl", hash = "sha256:2d9ad37fc657b1102ec880e637cccf20191581f75c64087a549e66c57e1ceb53", size = 1567424, upload-time = "2025-12-04T15:04:23.757Z" },
     { url = "https://files.pythonhosted.org/packages/05/f5/49a9ac2dff7f10091935def9165c90236d8f175afb27cbed38fb1d61ab6b/greenlet-3.3.0-cp312-cp312-musllinux_1_2_x86_64.whl", hash = "sha256:83cd0e36932e0e7f36a64b732a6f60c2fc2df28c351bae79fbaf4f8092fe7614", size = 1636017, upload-time = "2025-12-04T14:27:29.688Z" },
@@ -2034,6 +2037,7 @@ wheels = [
     { url = "https://files.pythonhosted.org/packages/02/2f/28592176381b9ab2cafa12829ba7b472d177f3acc35d8fbcf3673d966fff/greenlet-3.3.0-cp313-cp313-macosx_11_0_universal2.whl", hash = "sha256:a1e41a81c7e2825822f4e068c48cb2196002362619e2d70b148f20a831c00739", size = 275140, upload-time = "2025-12-04T14:23:01.282Z" },
     { url = "https://files.pythonhosted.org/packages/2c/80/fbe937bf81e9fca98c981fe499e59a3f45df2a04da0baa5c2be0dca0d329/greenlet-3.3.0-cp313-cp313-manylinux_2_24_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:9f515a47d02da4d30caaa85b69474cec77b7929b2e936ff7fb853d42f4bf8808", size = 599219, upload-time = "2025-12-04T14:50:08.309Z" },
     { url = "https://files.pythonhosted.org/packages/c2/ff/7c985128f0514271b8268476af89aee6866df5eec04ac17dcfbc676213df/greenlet-3.3.0-cp313-cp313-manylinux_2_24_ppc64le.manylinux_2_28_ppc64le.whl", hash = "sha256:7d2d9fd66bfadf230b385fdc90426fcd6eb64db54b40c495b72ac0feb5766c54", size = 610211, upload-time = "2025-12-04T14:57:43.968Z" },
+    { url = "https://files.pythonhosted.org/packages/79/07/c47a82d881319ec18a4510bb30463ed6891f2ad2c1901ed5ec23d3de351f/greenlet-3.3.0-cp313-cp313-manylinux_2_24_s390x.manylinux_2_28_s390x.whl", hash = "sha256:30a6e28487a790417d036088b3bcb3f3ac7d8babaa7d0139edbaddebf3af9492", size = 624311, upload-time = "2025-12-04T15:07:14.697Z" },
     { url = "https://files.pythonhosted.org/packages/fd/8e/424b8c6e78bd9837d14ff7df01a9829fc883ba2ab4ea787d4f848435f23f/greenlet-3.3.0-cp313-cp313-manylinux_2_24_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:087ea5e004437321508a8d6f20efc4cfec5e3c30118e1417ea96ed1d93950527", size = 612833, upload-time = "2025-12-04T14:26:03.669Z" },
     { url = "https://files.pythonhosted.org/packages/b5/ba/56699ff9b7c76ca12f1cdc27a886d0f81f2189c3455ff9f65246780f713d/greenlet-3.3.0-cp313-cp313-musllinux_1_2_aarch64.whl", hash = "sha256:ab97cf74045343f6c60a39913fa59710e4bd26a536ce7ab2397adf8b27e67c39", size = 1567256, upload-time = "2025-12-04T15:04:25.276Z" },
     { url = "https://files.pythonhosted.org/packages/1e/37/f31136132967982d698c71a281a8901daf1a8fbab935dce7c0cf15f942cc/greenlet-3.3.0-cp313-cp313-musllinux_1_2_x86_64.whl", hash = "sha256:5375d2e23184629112ca1ea89a53389dddbffcf417dad40125713d88eb5f96e8", size = 1636483, upload-time = "2025-12-04T14:27:30.804Z" },
@@ -2041,6 +2045,7 @@ wheels = [
     { url = "https://files.pythonhosted.org/packages/d7/7c/f0a6d0ede2c7bf092d00bc83ad5bafb7e6ec9b4aab2fbdfa6f134dc73327/greenlet-3.3.0-cp314-cp314-macosx_11_0_universal2.whl", hash = "sha256:60c2ef0f578afb3c8d92ea07ad327f9a062547137afe91f38408f08aacab667f", size = 275671, upload-time = "2025-12-04T14:23:05.267Z" },
     { url = "https://files.pythonhosted.org/packages/44/06/dac639ae1a50f5969d82d2e3dd9767d30d6dbdbab0e1a54010c8fe90263c/greenlet-3.3.0-cp314-cp314-manylinux_2_24_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:0a5d554d0712ba1de0a6c94c640f7aeba3f85b3a6e1f2899c11c2c0428da9365", size = 646360, upload-time = "2025-12-04T14:50:10.026Z" },
     { url = "https://files.pythonhosted.org/packages/e0/94/0fb76fe6c5369fba9bf98529ada6f4c3a1adf19e406a47332245ef0eb357/greenlet-3.3.0-cp314-cp314-manylinux_2_24_ppc64le.manylinux_2_28_ppc64le.whl", hash = "sha256:3a898b1e9c5f7307ebbde4102908e6cbfcb9ea16284a3abe15cab996bee8b9b3", size = 658160, upload-time = "2025-12-04T14:57:45.41Z" },
+    { url = "https://files.pythonhosted.org/packages/93/79/d2c70cae6e823fac36c3bbc9077962105052b7ef81db2f01ec3b9bf17e2b/greenlet-3.3.0-cp314-cp314-manylinux_2_24_s390x.manylinux_2_28_s390x.whl", hash = "sha256:dcd2bdbd444ff340e8d6bdf54d2f206ccddbb3ccfdcd3c25bf4afaa7b8f0cf45", size = 671388, upload-time = "2025-12-04T15:07:15.789Z" },
     { url = "https://files.pythonhosted.org/packages/b8/14/bab308fc2c1b5228c3224ec2bf928ce2e4d21d8046c161e44a2012b5203e/greenlet-3.3.0-cp314-cp314-manylinux_2_24_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:5773edda4dc00e173820722711d043799d3adb4f01731f40619e07ea2750b955", size = 660166, upload-time = "2025-12-04T14:26:05.099Z" },
     { url = "https://files.pythonhosted.org/packages/4b/d2/91465d39164eaa0085177f61983d80ffe746c5a1860f009811d498e7259c/greenlet-3.3.0-cp314-cp314-musllinux_1_2_aarch64.whl", hash = "sha256:ac0549373982b36d5fd5d30beb8a7a33ee541ff98d2b502714a09f1169f31b55", size = 1615193, upload-time = "2025-12-04T15:04:27.041Z" },
     { url = "https://files.pythonhosted.org/packages/42/1b/83d110a37044b92423084d52d5d5a3b3a73cafb51b547e6d7366ff62eff1/greenlet-3.3.0-cp314-cp314-musllinux_1_2_x86_64.whl", hash = "sha256:d198d2d977460358c3b3a4dc844f875d1adb33817f0613f663a656f463764ccc", size = 1683653, upload-time = "2025-12-04T14:27:32.366Z" },
@@ -2048,6 +2053,7 @@ wheels = [
     { url = "https://files.pythonhosted.org/packages/a0/66/bd6317bc5932accf351fc19f177ffba53712a202f9df10587da8df257c7e/greenlet-3.3.0-cp314-cp314t-macosx_11_0_universal2.whl", hash = "sha256:d6ed6f85fae6cdfdb9ce04c9bf7a08d666cfcfb914e7d006f44f840b46741931", size = 282638, upload-time = "2025-12-04T14:25:20.941Z" },
     { url = "https://files.pythonhosted.org/packages/30/cf/cc81cb030b40e738d6e69502ccbd0dd1bced0588e958f9e757945de24404/greenlet-3.3.0-cp314-cp314t-manylinux_2_24_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:d9125050fcf24554e69c4cacb086b87b3b55dc395a8b3ebe6487b045b2614388", size = 651145, upload-time = "2025-12-04T14:50:11.039Z" },
     { url = "https://files.pythonhosted.org/packages/9c/ea/1020037b5ecfe95ca7df8d8549959baceb8186031da83d5ecceff8b08cd2/greenlet-3.3.0-cp314-cp314t-manylinux_2_24_ppc64le.manylinux_2_28_ppc64le.whl", hash = "sha256:87e63ccfa13c0a0f6234ed0add552af24cc67dd886731f2261e46e241608bee3", size = 654236, upload-time = "2025-12-04T14:57:47.007Z" },
+    { url = "https://files.pythonhosted.org/packages/69/cc/1e4bae2e45ca2fa55299f4e85854606a78ecc37fead20d69322f96000504/greenlet-3.3.0-cp314-cp314t-manylinux_2_24_s390x.manylinux_2_28_s390x.whl", hash = "sha256:2662433acbca297c9153a4023fe2161c8dcfdcc91f10433171cf7e7d94ba2221", size = 662506, upload-time = "2025-12-04T15:07:16.906Z" },
     { url = "https://files.pythonhosted.org/packages/57/b9/f8025d71a6085c441a7eaff0fd928bbb275a6633773667023d19179fe815/greenlet-3.3.0-cp314-cp314t-manylinux_2_24_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:3c6e9b9c1527a78520357de498b0e709fb9e2f49c3a513afd5a249007261911b", size = 653783, upload-time = "2025-12-04T14:26:06.225Z" },
     { url = "https://files.pythonhosted.org/packages/f6/c7/876a8c7a7485d5d6b5c6821201d542ef28be645aa024cfe1145b35c120c1/greenlet-3.3.0-cp314-cp314t-musllinux_1_2_aarch64.whl", hash = "sha256:286d093f95ec98fdd92fcb955003b8a3d054b4e2cab3e2707a5039e7b50520fd", size = 1614857, upload-time = "2025-12-04T15:04:28.484Z" },
     { url = "https://files.pythonhosted.org/packages/4f/dc/041be1dff9f23dac5f48a43323cd0789cb798342011c19a248d9c9335536/greenlet-3.3.0-cp314-cp314t-musllinux_1_2_x86_64.whl", hash = "sha256:6c10513330af5b8ae16f023e8ddbfb486ab355d04467c4679c5cfe4659975dd9", size = 1676034, upload-time = "2025-12-04T14:27:33.531Z" },

From c89083e72ec405445ed47a644bcd5ff16ca00442 Mon Sep 17 00:00:00 2001
From: Paul Kompfner <paul@daily.co>
Date: Fri, 16 Jan 2026 22:26:31 -0500
Subject: [PATCH 11/51] Improve 20e example to ask the bot to give a recap when
 loading a previous conversation from disk

---
 .../foundational/20e-persistent-context-aws-nova-sonic.py | 8 ++++++++
 1 file changed, 8 insertions(+)

diff --git a/examples/foundational/20e-persistent-context-aws-nova-sonic.py b/examples/foundational/20e-persistent-context-aws-nova-sonic.py
index bd95fb7e7..ebd4afbf1 100644
--- a/examples/foundational/20e-persistent-context-aws-nova-sonic.py
+++ b/examples/foundational/20e-persistent-context-aws-nova-sonic.py
@@ -114,6 +114,14 @@ async def load_conversation(params: FunctionCallParams):
                 #         "content": f"{AWSNovaSonicLLMService.AWAIT_TRIGGER_ASSISTANT_RESPONSE_INSTRUCTION}",
                 #     }
                 # )
+                # If the last message isn't from the user, add a message asking for a recap
+                if messages and messages[-1].get("role") != "user":
+                    messages.append(
+                        {
+                            "role": "user",
+                            "content": "Can you catch me up on what we were talking about?",
+                        }
+                    )
                 params.context.set_messages(messages)
                 await params.llm.reset_conversation()
                 # await params.llm.trigger_assistant_response()

From b4d143e39b62c49542158cc392a3efa2702e7760 Mon Sep 17 00:00:00 2001
From: Paul Kompfner <paul@daily.co>
Date: Fri, 16 Jan 2026 22:36:35 -0500
Subject: [PATCH 12/51] Add CHANGELOG for fixing
 `AWSNovaSonicLLMService.reset_conversation()`

---
 changelog/3486.fixed.md | 1 +
 1 file changed, 1 insertion(+)
 create mode 100644 changelog/3486.fixed.md

diff --git a/changelog/3486.fixed.md b/changelog/3486.fixed.md
new file mode 100644
index 000000000..a02427e6e
--- /dev/null
+++ b/changelog/3486.fixed.md
@@ -0,0 +1 @@
+- Fixed `AWSNovaSonicLLMService.reset_conversation()`, which would previously error out. Now it successfully reconnects and "rehydrates" from the context object.

From 06b3ecd2d69d387a52a550672147e0d82468ba87 Mon Sep 17 00:00:00 2001
From: Paul Kompfner <paul@daily.co>
Date: Fri, 16 Jan 2026 22:43:49 -0500
Subject: [PATCH 13/51] In AWS Nova Sonic service, send the "interactive" user
 message (which triggers the bot response) only after sending the audio input
 start event, per the AWS team's recommendation

---
 src/pipecat/services/aws/nova_sonic/llm.py | 23 +++++++++++++++-------
 1 file changed, 16 insertions(+), 7 deletions(-)

diff --git a/src/pipecat/services/aws/nova_sonic/llm.py b/src/pipecat/services/aws/nova_sonic/llm.py
index e159ae9f6..05baba2bd 100644
--- a/src/pipecat/services/aws/nova_sonic/llm.py
+++ b/src/pipecat/services/aws/nova_sonic/llm.py
@@ -533,21 +533,30 @@ class AWSNovaSonicLLMService(LLMService):
         if system_instruction:
             await self._send_text_event(text=system_instruction, role=Role.SYSTEM)
 
-        # Send conversation history
+        # Send conversation history (except for the last message if it's from the
+        # user, which we'll send as interactive after starting audio input)
         messages = llm_connection_params["messages"]
+        last_user_message = None
         for i, message in enumerate(messages):
             # logger.debug(f"Seeding conversation history with message: {message}")
-            # If last message is from user, mark it as interactive to trigger
-            # bot response
             is_last_message = i == len(messages) - 1
-            interactive = is_last_message and message.role == Role.USER
-            await self._send_text_event(
-                text=message.text, role=message.role, interactive=interactive
-            )
+            if is_last_message and message.role == Role.USER:
+                # Save for sending after audio input starts
+                last_user_message = message
+            else:
+                await self._send_text_event(text=message.text, role=message.role)
 
         # Start audio input
         await self._send_audio_input_start_event()
 
+        # Now send the last user message as interactive to trigger bot response
+        if last_user_message:
+            # logger.debug(
+            #     f"Sending last user message as interactive to trigger bot response: {last_user_message}")
+            await self._send_text_event(
+                text=last_user_message.text, role=last_user_message.role, interactive=True
+            )
+
         # Start receiving events
         self._receive_task = self.create_task(self._receive_task_handler())
 

From 6cf0d53d0047d75eb164fa22564d8d0c84bfa2d9 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Aleix=20Conchillo=20Flaqu=C3=A9?= <aleix@daily.co>
Date: Mon, 19 Jan 2026 20:22:53 -0800
Subject: [PATCH 14/51] AIService: handle StartFrame/EndFrame/CancelFrame
 exceptions

If AIService subclasses implement start()/stop()/cancel() and exception are not
handled, execution will not continue and therefore the originator frames will
not be pushed. This would cause the pipeline to not be started (i.e. StartFrame
would not be pushed downstream) or stopped properly.
---
 changelog/3503.fixed.md            |  1 +
 src/pipecat/services/ai_service.py | 26 ++++++++++++++++++++++----
 2 files changed, 23 insertions(+), 4 deletions(-)
 create mode 100644 changelog/3503.fixed.md

diff --git a/changelog/3503.fixed.md b/changelog/3503.fixed.md
new file mode 100644
index 000000000..4218f8781
--- /dev/null
+++ b/changelog/3503.fixed.md
@@ -0,0 +1 @@
+- Fixed an issue in `AIService` where unhandled exceptions in `start()`, `stop()`, or `cancel()` implementations would prevent `process_frame()` to continue and therefore `StartFrame`, `EndFrame`, or `CancelFrame` from being pushed downstream, causing the pipeline to not start or stop properly.
diff --git a/src/pipecat/services/ai_service.py b/src/pipecat/services/ai_service.py
index a9952fa00..c03ab9d0e 100644
--- a/src/pipecat/services/ai_service.py
+++ b/src/pipecat/services/ai_service.py
@@ -148,11 +148,11 @@ class AIService(FrameProcessor):
         await super().process_frame(frame, direction)
 
         if isinstance(frame, StartFrame):
-            await self.start(frame)
-        elif isinstance(frame, CancelFrame):
-            await self.cancel(frame)
+            await self._start(frame)
         elif isinstance(frame, EndFrame):
-            await self.stop(frame)
+            await self._stop(frame)
+        elif isinstance(frame, CancelFrame):
+            await self._cancel(frame)
 
     async def process_generator(self, generator: AsyncGenerator[Frame | None, None]):
         """Process frames from an async generator.
@@ -169,3 +169,21 @@ class AIService(FrameProcessor):
                     await self.push_error_frame(f)
                 else:
                     await self.push_frame(f)
+
+    async def _start(self, frame: StartFrame):
+        try:
+            await self.start(frame)
+        except Exception as e:
+            logger.error(f"{self}: exception processing {frame}: {e}")
+
+    async def _stop(self, frame: EndFrame):
+        try:
+            await self.stop(frame)
+        except Exception as e:
+            logger.error(f"{self}: exception processing {frame}: {e}")
+
+    async def _cancel(self, frame: CancelFrame):
+        try:
+            await self.cancel(frame)
+        except Exception as e:
+            logger.error(f"{self}: exception processing {frame}: {e}")

From 9a718ded1ed2be52043cb23cd8b81ffb8bbaf223 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Aleix=20Conchillo=20Flaqu=C3=A9?= <aleix@daily.co>
Date: Mon, 19 Jan 2026 20:30:17 -0800
Subject: [PATCH 15/51] NvidiaTTSService: initialize client on StartFrame

Initialize client on StartFrame so errrors are reported within the pipeline.
---
 src/pipecat/services/nvidia/tts.py | 71 +++++++++++++++++++++---------
 1 file changed, 51 insertions(+), 20 deletions(-)

diff --git a/src/pipecat/services/nvidia/tts.py b/src/pipecat/services/nvidia/tts.py
index 70be24a8f..d62b7e500 100644
--- a/src/pipecat/services/nvidia/tts.py
+++ b/src/pipecat/services/nvidia/tts.py
@@ -25,6 +25,7 @@ from pydantic import BaseModel
 from pipecat.frames.frames import (
     ErrorFrame,
     Frame,
+    StartFrame,
     TTSAudioRawFrame,
     TTSStartedFrame,
     TTSStoppedFrame,
@@ -93,6 +94,7 @@ class NvidiaTTSService(TTSService):
 
         params = params or NvidiaTTSService.InputParams()
 
+        self._server = server
         self._api_key = api_key
         self._voice_id = voice_id
         self._language_code = params.language
@@ -102,18 +104,8 @@ class NvidiaTTSService(TTSService):
         self.set_model_name(model_function_map.get("model_name"))
         self.set_voice(voice_id)
 
-        metadata = [
-            ["function-id", self._function_id],
-            ["authorization", f"Bearer {api_key}"],
-        ]
-        auth = riva.client.Auth(None, self._use_ssl, server, metadata)
-
-        self._service = riva.client.SpeechSynthesisService(auth)
-
-        # warm up the service
-        config_response = self._service.stub.GetRivaSynthesisConfig(
-            riva.client.proto.riva_tts_pb2.RivaSynthesisConfigRequest()
-        )
+        self._service = None
+        self._config = None
 
     async def set_model(self, model: str):
         """Attempt to set the TTS model.
@@ -129,6 +121,39 @@ class NvidiaTTSService(TTSService):
             f"{self.__class__.__name__}(api_key=<api_key>, model_function_map={example})"
         )
 
+    def _initialize_client(self):
+        if self._service is not None:
+            return
+
+        metadata = [
+            ["function-id", self._function_id],
+            ["authorization", f"Bearer {self._api_key}"],
+        ]
+        auth = riva.client.Auth(None, self._use_ssl, self._server, metadata)
+
+        self._service = riva.client.SpeechSynthesisService(auth)
+
+    def _create_synthesis_config(self):
+        if not self._service:
+            return
+
+        # warm up the service
+        config = self._service.stub.GetRivaSynthesisConfig(
+            riva.client.proto.riva_tts_pb2.RivaSynthesisConfigRequest()
+        )
+        return config
+
+    async def start(self, frame: StartFrame):
+        """Start the Cartesia TTS service.
+
+        Args:
+            frame: The start frame containing initialization parameters.
+        """
+        await super().start(frame)
+        self._initialize_client()
+        self._config = self._create_synthesis_config()
+        logger.debug(f"Initialized NvidiaTTSService with model: {self.model_name}")
+
     @traced_tts
     async def run_tts(self, text: str) -> AsyncGenerator[Frame, None]:
         """Generate speech from text using NVIDIA Riva TTS.
@@ -161,12 +186,15 @@ class NvidiaTTSService(TTSService):
                 logger.error(f"{self} exception: {e}")
                 add_response(None)
 
-        await self.start_ttfb_metrics()
-        yield TTSStartedFrame()
-
-        logger.debug(f"{self}: Generating TTS [{text}]")
-
         try:
+            assert self._service is not None, "TTS service not initialized"
+            assert self._config is not None, "Synthesis configuration not created"
+
+            await self.start_ttfb_metrics()
+            yield TTSStartedFrame()
+
+            logger.debug(f"{self}: Generating TTS [{text}]")
+
             queue = asyncio.Queue()
             await asyncio.to_thread(read_audio_responses, queue)
 
@@ -181,9 +209,12 @@ class NvidiaTTSService(TTSService):
                 )
                 yield frame
                 resp = await asyncio.wait_for(queue.get(), timeout=NVIDIA_TTS_TIMEOUT_SECS)
+
+            await self.start_tts_usage_metrics(text)
+            yield TTSStoppedFrame()
         except asyncio.TimeoutError:
             logger.error(f"{self} timeout waiting for audio response")
             yield ErrorFrame(error=f"{self} error: {e}")
-
-        await self.start_tts_usage_metrics(text)
-        yield TTSStoppedFrame()
+        except Exception as e:
+            logger.error(f"{self} exception: {e}")
+            yield ErrorFrame(error=f"{self} error: {e}")

From 671dc8cd9bbe305c31f82d6891ff4a03cef0d6a4 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Aleix=20Conchillo=20Flaqu=C3=A9?= <aleix@daily.co>
Date: Mon, 19 Jan 2026 20:32:28 -0800
Subject: [PATCH 16/51] NvidiaSTTService: initialize client on StartFrame

Initialize client on StartFrame so errrors are reported within the pipeline.
---
 src/pipecat/services/nvidia/stt.py | 91 ++++++++++++++++--------------
 1 file changed, 49 insertions(+), 42 deletions(-)

diff --git a/src/pipecat/services/nvidia/stt.py b/src/pipecat/services/nvidia/stt.py
index 0d671571f..639e76535 100644
--- a/src/pipecat/services/nvidia/stt.py
+++ b/src/pipecat/services/nvidia/stt.py
@@ -134,6 +134,7 @@ class NvidiaSTTService(STTService):
 
         params = params or NvidiaSTTService.InputParams()
 
+        self._server = server
         self._api_key = api_key
         self._use_ssl = use_ssl
         self._profanity_filter = False
@@ -162,19 +163,55 @@ class NvidiaSTTService(STTService):
 
         self.set_model_name(model_function_map.get("model_name"))
 
-        metadata = [
-            ["function-id", self._function_id],
-            ["authorization", f"Bearer {api_key}"],
-        ]
-        auth = riva.client.Auth(None, self._use_ssl, server, metadata)
-
-        self._asr_service = riva.client.ASRService(auth)
-
+        self._asr_service = None
         self._queue = None
         self._config = None
         self._thread_task = None
         self._response_task = None
 
+    def _initialize_client(self):
+        metadata = [
+            ["function-id", self._function_id],
+            ["authorization", f"Bearer {self._api_key}"],
+        ]
+        auth = riva.client.Auth(None, self._use_ssl, self._server, metadata)
+
+        self._asr_service = riva.client.ASRService(auth)
+
+    def _create_recognition_config(self):
+        """Create the NVIDIA Riva ASR recognition configuration."""
+        config = riva.client.StreamingRecognitionConfig(
+            config=riva.client.RecognitionConfig(
+                encoding=riva.client.AudioEncoding.LINEAR_PCM,
+                language_code=self._language_code,
+                model="",
+                max_alternatives=1,
+                profanity_filter=self._profanity_filter,
+                enable_automatic_punctuation=self._automatic_punctuation,
+                verbatim_transcripts=not self._no_verbatim_transcripts,
+                sample_rate_hertz=self.sample_rate,
+                audio_channel_count=1,
+            ),
+            interim_results=True,
+        )
+
+        riva.client.add_word_boosting_to_config(
+            config, self._boosted_lm_words, self._boosted_lm_score
+        )
+
+        riva.client.add_endpoint_parameters_to_config(
+            config,
+            self._start_history,
+            self._start_threshold,
+            self._stop_history,
+            self._stop_history_eou,
+            self._stop_threshold,
+            self._stop_threshold_eou,
+        )
+        riva.client.add_custom_configuration_to_config(config, self._custom_configuration)
+
+        return config
+
     def can_generate_metrics(self) -> bool:
         """Check if this service can generate processing metrics.
 
@@ -206,41 +243,9 @@ class NvidiaSTTService(STTService):
             frame: StartFrame indicating pipeline start.
         """
         await super().start(frame)
+        self._initialize_client()
+        self._config = self._create_recognition_config()
 
-        if self._config:
-            return
-
-        config = riva.client.StreamingRecognitionConfig(
-            config=riva.client.RecognitionConfig(
-                encoding=riva.client.AudioEncoding.LINEAR_PCM,
-                language_code=self._language_code,
-                model="",
-                max_alternatives=1,
-                profanity_filter=self._profanity_filter,
-                enable_automatic_punctuation=self._automatic_punctuation,
-                verbatim_transcripts=not self._no_verbatim_transcripts,
-                sample_rate_hertz=self.sample_rate,
-                audio_channel_count=1,
-            ),
-            interim_results=True,
-        )
-
-        riva.client.add_word_boosting_to_config(
-            config, self._boosted_lm_words, self._boosted_lm_score
-        )
-
-        riva.client.add_endpoint_parameters_to_config(
-            config,
-            self._start_history,
-            self._start_threshold,
-            self._stop_history,
-            self._stop_history_eou,
-            self._stop_threshold,
-            self._stop_threshold_eou,
-        )
-        riva.client.add_custom_configuration_to_config(config, self._custom_configuration)
-
-        self._config = config
         self._queue = asyncio.Queue()
 
         if not self._thread_task:
@@ -250,6 +255,8 @@ class NvidiaSTTService(STTService):
             self._response_queue = asyncio.Queue()
             self._response_task = self.create_task(self._response_task_handler())
 
+        logger.debug(f"Initialized NvidiaSTTService with model: {self.model_name}")
+
     async def stop(self, frame: EndFrame):
         """Stop the NVIDIA Riva STT service and clean up resources.
 

From 655006aff5acccccdda50a3f01c990e844b7843b Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Aleix=20Conchillo=20Flaqu=C3=A9?= <aleix@daily.co>
Date: Mon, 19 Jan 2026 20:31:58 -0800
Subject: [PATCH 17/51] NvidiaSegmentedSTTService: simplify exception handling

---
 src/pipecat/services/nvidia/stt.py | 79 +++++++++++++-----------------
 1 file changed, 33 insertions(+), 46 deletions(-)

diff --git a/src/pipecat/services/nvidia/stt.py b/src/pipecat/services/nvidia/stt.py
index 639e76535..45c59de7a 100644
--- a/src/pipecat/services/nvidia/stt.py
+++ b/src/pipecat/services/nvidia/stt.py
@@ -510,8 +510,6 @@ class NvidiaSegmentedSTTService(SegmentedSTTService):
         auth = riva.client.Auth(None, self._use_ssl, self._server, metadata)
         self._asr_service = riva.client.ASRService(auth)
 
-        logger.info(f"Initialized NvidiaSegmentedSTTService with model: {self.model_name}")
-
     def _create_recognition_config(self):
         """Create the NVIDIA Riva ASR recognition configuration."""
         # Create base configuration
@@ -579,6 +577,7 @@ class NvidiaSegmentedSTTService(SegmentedSTTService):
         await super().start(frame)
         self._initialize_client()
         self._config = self._create_recognition_config()
+        logger.debug(f"Initialized NvidiaSegmentedSTTService with model: {self.model_name}")
 
     async def set_language(self, language: Language):
         """Set the language for the STT service.
@@ -612,21 +611,12 @@ class NvidiaSegmentedSTTService(SegmentedSTTService):
             Frame: TranscriptionFrame containing the transcribed text.
         """
         try:
-            await self.start_processing_metrics()
-            await self.start_ttfb_metrics()
-
-            # Make sure the client is initialized
-            if self._asr_service is None:
-                self._initialize_client()
-
-            # Make sure the config is created
-            if self._config is None:
-                self._config = self._create_recognition_config()
-
-            # Type assertion to satisfy the IDE
             assert self._asr_service is not None, "ASR service not initialized"
             assert self._config is not None, "Recognition config not created"
 
+            await self.start_processing_metrics()
+            await self.start_ttfb_metrics()
+
             # Process audio with NVIDIA Riva ASR - explicitly request non-future response
             raw_response = self._asr_service.offline_recognize(audio, self._config, future=False)
 
@@ -634,43 +624,40 @@ class NvidiaSegmentedSTTService(SegmentedSTTService):
             await self.stop_processing_metrics()
 
             # Process the response - handle different possible return types
-            try:
-                # If it's a future-like object, get the result
-                if hasattr(raw_response, "result"):
-                    response = raw_response.result()
-                else:
-                    response = raw_response
+            # If it's a future-like object, get the result
+            if hasattr(raw_response, "result"):
+                response = raw_response.result()
+            else:
+                response = raw_response
 
-                # Process transcription results
-                transcription_found = False
+            # Process transcription results
+            transcription_found = False
 
-                # Now we can safely check results
-                # Type hint for the IDE
-                results = getattr(response, "results", [])
+            # Now we can safely check results
+            # Type hint for the IDE
+            results = getattr(response, "results", [])
 
-                for result in results:
-                    alternatives = getattr(result, "alternatives", [])
-                    if alternatives:
-                        text = alternatives[0].transcript.strip()
-                        if text:
-                            logger.debug(f"Transcription: [{text}]")
-                            yield TranscriptionFrame(
-                                text,
-                                self._user_id,
-                                time_now_iso8601(),
-                                self._language_enum,
-                            )
-                            transcription_found = True
+            for result in results:
+                alternatives = getattr(result, "alternatives", [])
+                if alternatives:
+                    text = alternatives[0].transcript.strip()
+                    if text:
+                        logger.debug(f"Transcription: [{text}]")
+                        yield TranscriptionFrame(
+                            text,
+                            self._user_id,
+                            time_now_iso8601(),
+                            self._language_enum,
+                        )
+                        transcription_found = True
 
-                            await self._handle_transcription(text, True, self._language_enum)
-
-                if not transcription_found:
-                    logger.debug("No transcription results found in NVIDIA Riva response")
-
-            except AttributeError as ae:
-                logger.error(f"Unexpected response structure from NVIDIA Riva: {ae}")
-                yield ErrorFrame(f"Unexpected NVIDIA Riva response format: {str(ae)}")
+                        await self._handle_transcription(text, True, self._language_enum)
 
+            if not transcription_found:
+                logger.debug(f"{self}: No transcription results found in NVIDIA Riva response")
+        except AttributeError as ae:
+            logger.error(f"{self}: Unexpected response structure from NVIDIA Riva: {ae}")
+            yield ErrorFrame(f"{self}: Unexpected NVIDIA Riva response format: {str(ae)}")
         except Exception as e:
             logger.error(f"{self} exception: {e}")
             yield ErrorFrame(error=f"{self} error: {e}")

From a010a020fd7db01a660740de852efe45baf70970 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Aleix=20Conchillo=20Flaqu=C3=A9?= <aleix@daily.co>
Date: Tue, 20 Jan 2026 09:03:30 -0800
Subject: [PATCH 18/51] add changelog fo 3504

---
 changelog/3504.fixed.md | 1 +
 1 file changed, 1 insertion(+)
 create mode 100644 changelog/3504.fixed.md

diff --git a/changelog/3504.fixed.md b/changelog/3504.fixed.md
new file mode 100644
index 000000000..2d675547a
--- /dev/null
+++ b/changelog/3504.fixed.md
@@ -0,0 +1 @@
+- Moved `NVIDIATTSService` and `NVIDIASTTService` client initialization from constructor to `start()` for better error handling.

From fa6f924b31b086942942947589d8fe222e2bc260 Mon Sep 17 00:00:00 2001
From: Sunah Suh <sunah@daily.co>
Date: Fri, 16 Jan 2026 13:51:00 -0600
Subject: [PATCH 19/51] Log Daily participant and meeting session IDs upon
 successful join in Daily Transport

---
 src/pipecat/transports/daily/transport.py | 6 +++++-
 1 file changed, 5 insertions(+), 1 deletion(-)

diff --git a/src/pipecat/transports/daily/transport.py b/src/pipecat/transports/daily/transport.py
index 203cbb9fa..3b652b517 100644
--- a/src/pipecat/transports/daily/transport.py
+++ b/src/pipecat/transports/daily/transport.py
@@ -759,7 +759,11 @@ class DailyTransportClient(EventHandler):
             # Increment leave counter if we successfully joined.
             self._leave_counter += 1
 
-            logger.info(f"Joined {self._room_url}")
+            participant_id = data.get("participants", {}).get("local", {}).get("id")
+            meeting_id = data.get("meetingSession", {}).get("id")
+            logger.info(
+                f"Joined {self._room_url}. Participant ID: {participant_id}, Meeting ID: {meeting_id}"
+            )
 
             await self._callbacks.on_joined(data)
 

From 461bd0a2e089780e11888c276ee89392bc614ce9 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Aleix=20Conchillo=20Flaqu=C3=A9?= <aleix@daily.co>
Date: Tue, 20 Jan 2026 13:26:40 -0800
Subject: [PATCH 20/51] update changelog for #3494 and #3499

---
 changelog/3494.fixed.md | 1 +
 changelog/3499.fixed.md | 1 +
 2 files changed, 2 insertions(+)
 create mode 100644 changelog/3494.fixed.md
 create mode 100644 changelog/3499.fixed.md

diff --git a/changelog/3494.fixed.md b/changelog/3494.fixed.md
new file mode 100644
index 000000000..41fbfa74e
--- /dev/null
+++ b/changelog/3494.fixed.md
@@ -0,0 +1 @@
+- Fixed an issue where `UninterruptibleFrame` frames would not be preserved in some cases.
diff --git a/changelog/3499.fixed.md b/changelog/3499.fixed.md
new file mode 100644
index 000000000..ae893c13d
--- /dev/null
+++ b/changelog/3499.fixed.md
@@ -0,0 +1 @@
+- Fixed memory leak in `LiveKitTransport` when `video_in_enabled` is `False`.

From 14495c425a42c6b9ca1c9ec4a55429388d10e502 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Aleix=20Conchillo=20Flaqu=C3=A9?= <aleix@daily.co>
Date: Tue, 20 Jan 2026 13:43:23 -0800
Subject: [PATCH 21/51] NVIDIASTTService: no need for additional queue and task

---
 changelog/3509.fixed.md            |  1 +
 src/pipecat/services/nvidia/stt.py | 19 +------------------
 2 files changed, 2 insertions(+), 18 deletions(-)
 create mode 100644 changelog/3509.fixed.md

diff --git a/changelog/3509.fixed.md b/changelog/3509.fixed.md
new file mode 100644
index 000000000..153060b46
--- /dev/null
+++ b/changelog/3509.fixed.md
@@ -0,0 +1 @@
+- Optimized `NVIDIASTTService` by removing unnecessary queue and task.
diff --git a/src/pipecat/services/nvidia/stt.py b/src/pipecat/services/nvidia/stt.py
index 45c59de7a..7d52f5130 100644
--- a/src/pipecat/services/nvidia/stt.py
+++ b/src/pipecat/services/nvidia/stt.py
@@ -167,7 +167,6 @@ class NvidiaSTTService(STTService):
         self._queue = None
         self._config = None
         self._thread_task = None
-        self._response_task = None
 
     def _initialize_client(self):
         metadata = [
@@ -251,10 +250,6 @@ class NvidiaSTTService(STTService):
         if not self._thread_task:
             self._thread_task = self.create_task(self._thread_task_handler())
 
-        if not self._response_task:
-            self._response_queue = asyncio.Queue()
-            self._response_task = self.create_task(self._response_task_handler())
-
         logger.debug(f"Initialized NvidiaSTTService with model: {self.model_name}")
 
     async def stop(self, frame: EndFrame):
@@ -280,10 +275,6 @@ class NvidiaSTTService(STTService):
             await self.cancel_task(self._thread_task)
             self._thread_task = None
 
-        if self._response_task:
-            await self.cancel_task(self._response_task)
-            self._response_task = None
-
     def _response_handler(self):
         responses = self._asr_service.streaming_response_generator(
             audio_chunks=self,
@@ -292,9 +283,7 @@ class NvidiaSTTService(STTService):
         for response in responses:
             if not response.results:
                 continue
-            asyncio.run_coroutine_threadsafe(
-                self._response_queue.put(response), self.get_event_loop()
-            )
+            asyncio.run_coroutine_threadsafe(self._handle_response(response), self.get_event_loop())
 
     async def _thread_task_handler(self):
         try:
@@ -346,12 +335,6 @@ class NvidiaSTTService(STTService):
                         )
                     )
 
-    async def _response_task_handler(self):
-        while True:
-            response = await self._response_queue.get()
-            await self._handle_response(response)
-            self._response_queue.task_done()
-
     async def run_stt(self, audio: bytes) -> AsyncGenerator[Frame, None]:
         """Process audio data for speech-to-text transcription.
 

From a787fd9cd81c1518df1bbd26909d5e2536c47f20 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Aleix=20Conchillo=20Flaqu=C3=A9?= <aleix@daily.co>
Date: Tue, 20 Jan 2026 13:44:52 -0800
Subject: [PATCH 22/51] NVIDIATTSService: process incoming audio frame right
 away

Process audio as soon as we receive it from the generator. Previously, we were
reading from the generator and adding elements into a queue until there was no
more data, then we would process the queue.
---
 changelog/3509.fixed.2.md          |  1 +
 src/pipecat/services/nvidia/tts.py | 54 ++++++++++++++----------------
 2 files changed, 27 insertions(+), 28 deletions(-)
 create mode 100644 changelog/3509.fixed.2.md

diff --git a/changelog/3509.fixed.2.md b/changelog/3509.fixed.2.md
new file mode 100644
index 000000000..68053011b
--- /dev/null
+++ b/changelog/3509.fixed.2.md
@@ -0,0 +1 @@
+- Optimized `NVIDIATTSService` to process incoming audio frames immediately.
diff --git a/src/pipecat/services/nvidia/tts.py b/src/pipecat/services/nvidia/tts.py
index d62b7e500..eddafce01 100644
--- a/src/pipecat/services/nvidia/tts.py
+++ b/src/pipecat/services/nvidia/tts.py
@@ -12,7 +12,7 @@ gRPC API for high-quality speech synthesis.
 
 import asyncio
 import os
-from typing import AsyncGenerator, Mapping, Optional
+from typing import AsyncGenerator, AsyncIterable, Generator, Mapping, Optional
 
 from pipecat.utils.tracing.service_decorators import traced_tts
 
@@ -35,14 +35,12 @@ from pipecat.transcriptions.language import Language
 
 try:
     import riva.client
-
+    import riva.client.proto.riva_tts_pb2 as rtts
 except ModuleNotFoundError as e:
     logger.error(f"Exception: {e}")
     logger.error("In order to use NVIDIA Riva TTS, you need to `pip install pipecat-ai[nvidia]`.")
     raise Exception(f"Missing module: {e}")
 
-NVIDIA_TTS_TIMEOUT_SECS = 5
-
 
 class NvidiaTTSService(TTSService):
     """NVIDIA Riva text-to-speech service.
@@ -165,26 +163,30 @@ class NvidiaTTSService(TTSService):
             Frame: Audio frames containing the synthesized speech data.
         """
 
-        def read_audio_responses(queue: asyncio.Queue):
-            def add_response(r):
-                asyncio.run_coroutine_threadsafe(queue.put(r), self.get_event_loop())
+        def read_audio_responses() -> Generator[rtts.SynthesizeSpeechResponse, None, None]:
+            responses = self._service.synthesize_online(
+                text,
+                self._voice_id,
+                self._language_code,
+                sample_rate_hz=self.sample_rate,
+                zero_shot_audio_prompt_file=None,
+                zero_shot_quality=self._quality,
+                custom_dictionary={},
+            )
+            return responses
 
+        def async_next(it):
             try:
-                responses = self._service.synthesize_online(
-                    text,
-                    self._voice_id,
-                    self._language_code,
-                    sample_rate_hz=self.sample_rate,
-                    zero_shot_audio_prompt_file=None,
-                    zero_shot_quality=self._quality,
-                    custom_dictionary={},
-                )
-                for r in responses:
-                    add_response(r)
-                add_response(None)
-            except Exception as e:
-                logger.error(f"{self} exception: {e}")
-                add_response(None)
+                return next(it)
+            except StopIteration:
+                return None
+
+        async def async_iterator(iterator) -> AsyncIterable[rtts.SynthesizeSpeechResponse]:
+            while True:
+                item = await asyncio.to_thread(async_next, iterator)
+                if item is None:
+                    return
+                yield item
 
         try:
             assert self._service is not None, "TTS service not initialized"
@@ -195,12 +197,9 @@ class NvidiaTTSService(TTSService):
 
             logger.debug(f"{self}: Generating TTS [{text}]")
 
-            queue = asyncio.Queue()
-            await asyncio.to_thread(read_audio_responses, queue)
+            responses = await asyncio.to_thread(read_audio_responses)
 
-            # Wait for the thread to start.
-            resp = await asyncio.wait_for(queue.get(), timeout=NVIDIA_TTS_TIMEOUT_SECS)
-            while resp:
+            async for resp in async_iterator(responses):
                 await self.stop_ttfb_metrics()
                 frame = TTSAudioRawFrame(
                     audio=resp.audio,
@@ -208,7 +207,6 @@ class NvidiaTTSService(TTSService):
                     num_channels=1,
                 )
                 yield frame
-                resp = await asyncio.wait_for(queue.get(), timeout=NVIDIA_TTS_TIMEOUT_SECS)
 
             await self.start_tts_usage_metrics(text)
             yield TTSStoppedFrame()

From 7e0ca113afa73adffce9bc50354d3b6b51db5dda Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Aleix=20Conchillo=20Flaqu=C3=A9?= <aleix@daily.co>
Date: Tue, 20 Jan 2026 19:04:58 -0800
Subject: [PATCH 23/51] CambTTSService: initialize client during StartFrame

---
 changelog/3511.fixed.md          |  1 +
 src/pipecat/services/camb/tts.py | 11 +++++++++--
 2 files changed, 10 insertions(+), 2 deletions(-)
 create mode 100644 changelog/3511.fixed.md

diff --git a/changelog/3511.fixed.md b/changelog/3511.fixed.md
new file mode 100644
index 000000000..1f4f429ac
--- /dev/null
+++ b/changelog/3511.fixed.md
@@ -0,0 +1 @@
+- Fixed a `CambTTSService` issue where client was being initialized in the constructor which wouldn't allow for proper Pipeline error handling.
diff --git a/src/pipecat/services/camb/tts.py b/src/pipecat/services/camb/tts.py
index 89279604d..00b4eaf9b 100644
--- a/src/pipecat/services/camb/tts.py
+++ b/src/pipecat/services/camb/tts.py
@@ -199,9 +199,10 @@ class CambTTSService(TTSService):
         """
         super().__init__(sample_rate=sample_rate, **kwargs)
 
-        params = params or CambTTSService.InputParams()
+        self._api_key = api_key
+        self._timeout = timeout
 
-        self._client = AsyncCambAI(api_key=api_key, timeout=timeout)
+        params = params or CambTTSService.InputParams()
 
         # Warn if sample rate doesn't match model's supported rate
         if sample_rate and sample_rate != MODEL_SAMPLE_RATES.get(model):
@@ -222,6 +223,8 @@ class CambTTSService(TTSService):
         self.set_voice(str(voice_id))
         self._voice_id = voice_id
 
+        self._client = None
+
     def can_generate_metrics(self) -> bool:
         """Check if this service can generate processing metrics.
 
@@ -249,6 +252,8 @@ class CambTTSService(TTSService):
         """
         await super().start(frame)
 
+        self._client = AsyncCambAI(api_key=self._api_key, timeout=self._timeout)
+
         # Use model-specific sample rate if not explicitly specified
         if not self._init_sample_rate:
             self._sample_rate = MODEL_SAMPLE_RATES.get(self.model_name, 22050)
@@ -289,6 +294,8 @@ class CambTTSService(TTSService):
             await self.start_tts_usage_metrics(text)
             yield TTSStartedFrame()
 
+            assert self._client is not None, "Camb.ai TTS service not initialized"
+
             # Buffer for aligning chunks to 2-byte boundaries (16-bit PCM)
             audio_buffer = b""
 

From 5f9ff8bd58341bec4720ed0264fbdce03b66db0f Mon Sep 17 00:00:00 2001
From: aconchillo <951761+aconchillo@users.noreply.github.com>
Date: Wed, 21 Jan 2026 03:18:50 +0000
Subject: [PATCH 24/51] Update changelog for version 0.0.100

---
 CHANGELOG.md                 | 123 +++++++++++++++++++++++++++++++++++
 changelog/3169.added.md      |   1 -
 changelog/3287.changed.md    |   1 -
 changelog/3287.fixed.md      |   1 -
 changelog/3349.added.md      |   1 -
 changelog/3446.fixed.md      |   8 ---
 changelog/3454.fixed.md      |   1 -
 changelog/3455.fixed.md      |   1 -
 changelog/3461.added.md      |   1 -
 changelog/3462.fixed.md      |   1 -
 changelog/3479.deprecated.md |   1 -
 changelog/3480.fixed.md      |   1 -
 changelog/3482.added.md      |   1 -
 changelog/3483.changed.md    |   1 -
 changelog/3484.fixed.md      |   1 -
 changelog/3486.fixed.md      |   1 -
 changelog/3489.fixed.md      |   3 -
 changelog/3490.added.md      |   1 -
 changelog/3494.fixed.md      |   1 -
 changelog/3499.fixed.md      |   1 -
 changelog/3503.fixed.md      |   1 -
 changelog/3504.fixed.md      |   1 -
 changelog/3509.fixed.2.md    |   1 -
 changelog/3509.fixed.md      |   1 -
 changelog/3511.fixed.md      |   1 -
 25 files changed, 123 insertions(+), 33 deletions(-)
 delete mode 100644 changelog/3169.added.md
 delete mode 100644 changelog/3287.changed.md
 delete mode 100644 changelog/3287.fixed.md
 delete mode 100644 changelog/3349.added.md
 delete mode 100644 changelog/3446.fixed.md
 delete mode 100644 changelog/3454.fixed.md
 delete mode 100644 changelog/3455.fixed.md
 delete mode 100644 changelog/3461.added.md
 delete mode 100644 changelog/3462.fixed.md
 delete mode 100644 changelog/3479.deprecated.md
 delete mode 100644 changelog/3480.fixed.md
 delete mode 100644 changelog/3482.added.md
 delete mode 100644 changelog/3483.changed.md
 delete mode 100644 changelog/3484.fixed.md
 delete mode 100644 changelog/3486.fixed.md
 delete mode 100644 changelog/3489.fixed.md
 delete mode 100644 changelog/3490.added.md
 delete mode 100644 changelog/3494.fixed.md
 delete mode 100644 changelog/3499.fixed.md
 delete mode 100644 changelog/3503.fixed.md
 delete mode 100644 changelog/3504.fixed.md
 delete mode 100644 changelog/3509.fixed.2.md
 delete mode 100644 changelog/3509.fixed.md
 delete mode 100644 changelog/3511.fixed.md

diff --git a/CHANGELOG.md b/CHANGELOG.md
index ce2bdef9f..1d70f0682 100644
--- a/CHANGELOG.md
+++ b/CHANGELOG.md
@@ -7,6 +7,129 @@ and this project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0
 
 <!-- towncrier release notes start -->
 
+## [0.0.100] - 2026-01-20
+
+### Added
+
+- Added Hathora service to support Hathora-hosted TTS and STT models (only
+  non-streaming)
+  (PR [#3169](https://github.com/pipecat-ai/pipecat/pull/3169))
+
+- Added `CambTTSService`, using Camb.ai's TTS integration with MARS models
+  (mars-flash, mars-pro, mars-instruct) for high-quality text-to-speech
+  synthesis.
+  (PR [#3349](https://github.com/pipecat-ai/pipecat/pull/3349))
+
+- Added the `additional_headers` param to `WebsocketClientParams`, allowing
+  `WebsocketClientTransport` to send custom headers on connect, for cases such
+  as authentication.
+  (PR [#3461](https://github.com/pipecat-ai/pipecat/pull/3461))
+
+- Added `UserIdleController` for detecting user idle state, integrated into
+  `LLMUserAggregator` and `UserTurnProcessor` via optional `user_idle_timeout`
+  parameter. Emits `on_user_turn_idle` event for application-level handling.
+  Deprecated `UserIdleProcessor` in favor of the new compositional approach.
+  (PR [#3482](https://github.com/pipecat-ai/pipecat/pull/3482))
+
+- Added `on_user_mute_started` and `on_user_mute_stopped` event handlers to
+  `LLMUserAggregator` for tracking user mute state changes.
+  (PR [#3490](https://github.com/pipecat-ai/pipecat/pull/3490))
+
+### Changed
+
+- Enhanced interruption handling in `AsyncAITTSService` by supporting
+  multi-context WebSocket sessions for more robust context management.
+  (PR [#3287](https://github.com/pipecat-ai/pipecat/pull/3287))
+
+- Throttle `UserSpeakingFrame` to broadcast at most every 200ms instead of on
+  every audio chunk, reducing frame processing overhead during user speech.
+  (PR [#3483](https://github.com/pipecat-ai/pipecat/pull/3483))
+
+### Deprecated
+
+- For consistency with other package names, we just deprecated
+  `pipecat.turns.mute` (introduced in Pipecat 0.0.99) in favor of
+  `pipecat.turns.user_mute`.
+  (PR [#3479](https://github.com/pipecat-ai/pipecat/pull/3479))
+
+### Fixed
+
+- Corrected TTFB metric calculation in `AsyncAIHttpTTSService`.
+  (PR [#3287](https://github.com/pipecat-ai/pipecat/pull/3287))
+
+- Fixed an issue where the "bot-llm-text" RTVI event would not fire for
+  realtime (speech-to-speech) services:
+
+    - `AWSNovaSonicLLMService`
+    - `GeminiLiveLLMService`
+    - `OpenAIRealtimeLLMService`
+    - `GrokRealtimeLLMService`
+
+  The issue was that these services weren't pushing `LLMTextFrame`s. Now
+  they  do.
+  (PR [#3446](https://github.com/pipecat-ai/pipecat/pull/3446))
+
+- Fixed an issue where `on_user_turn_stop_timeout` could fire while a user is
+  talking when using `ExternalUserTurnStrategies`.
+  (PR [#3454](https://github.com/pipecat-ai/pipecat/pull/3454))
+
+- Fixed an issue where user turn start strategies were not being reset after a
+  user turn started, causing incorrect strategy behavior.
+  (PR [#3455](https://github.com/pipecat-ai/pipecat/pull/3455))
+
+- Fixed `MinWordsUserTurnStartStrategy` to not aggregate transcriptions,
+  preventing incorrect turn starts when words are spoken with pauses between
+  them.
+  (PR [#3462](https://github.com/pipecat-ai/pipecat/pull/3462))
+
+- Fixed an issue where Grok Realtime would error out when running with
+  SmallWebRTC transport.
+  (PR [#3480](https://github.com/pipecat-ai/pipecat/pull/3480))
+
+- Fixed a `Mem0MemoryService` issue where passing `async_mode: true` was
+  causing an error. See
+  https://docs.mem0.ai/platform/features/async-mode-default-change.
+  (PR [#3484](https://github.com/pipecat-ai/pipecat/pull/3484))
+
+- Fixed `AWSNovaSonicLLMService.reset_conversation()`, which would previously
+  error out. Now it successfully reconnects and "rehydrates" from the context
+  object.
+  (PR [#3486](https://github.com/pipecat-ai/pipecat/pull/3486))
+
+- Fixed `AzureTTSService` transcript formatting issues:
+    - Punctuation now appears without extra spaces (e.g., "Hello!" instead of
+      "Hello !")
+    - CJK languages (Chinese, Japanese, Korean) no longer have unwanted spaces
+      between characters
+  (PR [#3489](https://github.com/pipecat-ai/pipecat/pull/3489))
+
+- Fixed an issue where `UninterruptibleFrame` frames would not be preserved in
+  some cases.
+  (PR [#3494](https://github.com/pipecat-ai/pipecat/pull/3494))
+
+- Fixed memory leak in `LiveKitTransport` when `video_in_enabled` is `False`.
+  (PR [#3499](https://github.com/pipecat-ai/pipecat/pull/3499))
+
+- Fixed an issue in `AIService` where unhandled exceptions in `start()`,
+  `stop()`, or `cancel()` implementations would prevent `process_frame()` to
+  continue and therefore `StartFrame`, `EndFrame`, or `CancelFrame` from being
+  pushed downstream, causing the pipeline to not start or stop properly.
+  (PR [#3503](https://github.com/pipecat-ai/pipecat/pull/3503))
+
+- Moved `NVIDIATTSService` and `NVIDIASTTService` client initialization from
+  constructor to `start()` for better error handling.
+  (PR [#3504](https://github.com/pipecat-ai/pipecat/pull/3504))
+
+- Optimized `NVIDIATTSService` to process incoming audio frames immediately.
+  (PR [#3509](https://github.com/pipecat-ai/pipecat/pull/3509))
+
+- Optimized `NVIDIASTTService` by removing unnecessary queue and task.
+  (PR [#3509](https://github.com/pipecat-ai/pipecat/pull/3509))
+
+- Fixed a `CambTTSService` issue where client was being initialized in the
+  constructor which wouldn't allow for proper Pipeline error handling.
+  (PR [#3511](https://github.com/pipecat-ai/pipecat/pull/3511))
+
 ## [0.0.99] - 2026-01-13
 
 ### Added
diff --git a/changelog/3169.added.md b/changelog/3169.added.md
deleted file mode 100644
index d9e4bb9b9..000000000
--- a/changelog/3169.added.md
+++ /dev/null
@@ -1 +0,0 @@
-- Added Hathora service to support Hathora-hosted TTS and STT models (only non-streaming)
\ No newline at end of file
diff --git a/changelog/3287.changed.md b/changelog/3287.changed.md
deleted file mode 100644
index f0df82966..000000000
--- a/changelog/3287.changed.md
+++ /dev/null
@@ -1 +0,0 @@
-- Enhanced interruption handling in `AsyncAITTSService` by supporting multi-context WebSocket sessions for more robust context management.
\ No newline at end of file
diff --git a/changelog/3287.fixed.md b/changelog/3287.fixed.md
deleted file mode 100644
index 30ce0b13b..000000000
--- a/changelog/3287.fixed.md
+++ /dev/null
@@ -1 +0,0 @@
-- Corrected TTFB metric calculation in `AsyncAIHttpTTSService`.
\ No newline at end of file
diff --git a/changelog/3349.added.md b/changelog/3349.added.md
deleted file mode 100644
index a5b282ec8..000000000
--- a/changelog/3349.added.md
+++ /dev/null
@@ -1 +0,0 @@
-- Added `CambTTSService`, using Camb.ai's TTS integration with MARS models (mars-flash, mars-pro, mars-instruct) for high-quality text-to-speech synthesis.
diff --git a/changelog/3446.fixed.md b/changelog/3446.fixed.md
deleted file mode 100644
index 64cc3cb32..000000000
--- a/changelog/3446.fixed.md
+++ /dev/null
@@ -1,8 +0,0 @@
-- Fixed an issue where the "bot-llm-text" RTVI event would not fire for realtime (speech-to-speech) services:
-
-  - `AWSNovaSonicLLMService`
-  - `GeminiLiveLLMService`
-  - `OpenAIRealtimeLLMService`
-  - `GrokRealtimeLLMService`
-
-  The issue was that these services weren't pushing `LLMTextFrame`s. Now they do.
diff --git a/changelog/3454.fixed.md b/changelog/3454.fixed.md
deleted file mode 100644
index 0269370d7..000000000
--- a/changelog/3454.fixed.md
+++ /dev/null
@@ -1 +0,0 @@
-- Fixed an issue where `on_user_turn_stop_timeout` could fire while a user is talking when using `ExternalUserTurnStrategies`.
diff --git a/changelog/3455.fixed.md b/changelog/3455.fixed.md
deleted file mode 100644
index 1dba5838a..000000000
--- a/changelog/3455.fixed.md
+++ /dev/null
@@ -1 +0,0 @@
-- Fixed an issue where user turn start strategies were not being reset after a user turn started, causing incorrect strategy behavior.
diff --git a/changelog/3461.added.md b/changelog/3461.added.md
deleted file mode 100644
index e0bc27d83..000000000
--- a/changelog/3461.added.md
+++ /dev/null
@@ -1 +0,0 @@
-- Added the `additional_headers` param to `WebsocketClientParams`, allowing `WebsocketClientTransport` to send custom headers on connect, for cases such as authentication.
diff --git a/changelog/3462.fixed.md b/changelog/3462.fixed.md
deleted file mode 100644
index f9ede6a53..000000000
--- a/changelog/3462.fixed.md
+++ /dev/null
@@ -1 +0,0 @@
-- Fixed `MinWordsUserTurnStartStrategy` to not aggregate transcriptions, preventing incorrect turn starts when words are spoken with pauses between them.
diff --git a/changelog/3479.deprecated.md b/changelog/3479.deprecated.md
deleted file mode 100644
index 7c58a9d89..000000000
--- a/changelog/3479.deprecated.md
+++ /dev/null
@@ -1 +0,0 @@
-- For consistency with other package names, we just deprecated `pipecat.turns.mute` (introduced in Pipecat 0.0.99) in favor of `pipecat.turns.user_mute`.
diff --git a/changelog/3480.fixed.md b/changelog/3480.fixed.md
deleted file mode 100644
index 9d04545ec..000000000
--- a/changelog/3480.fixed.md
+++ /dev/null
@@ -1 +0,0 @@
-- Fixed an issue where Grok Realtime would error out when running with SmallWebRTC transport.
diff --git a/changelog/3482.added.md b/changelog/3482.added.md
deleted file mode 100644
index 465d409f4..000000000
--- a/changelog/3482.added.md
+++ /dev/null
@@ -1 +0,0 @@
-- Added `UserIdleController` for detecting user idle state, integrated into `LLMUserAggregator` and `UserTurnProcessor` via optional `user_idle_timeout` parameter. Emits `on_user_turn_idle` event for application-level handling. Deprecated `UserIdleProcessor` in favor of the new compositional approach.
diff --git a/changelog/3483.changed.md b/changelog/3483.changed.md
deleted file mode 100644
index 97bb10371..000000000
--- a/changelog/3483.changed.md
+++ /dev/null
@@ -1 +0,0 @@
-- Throttle `UserSpeakingFrame` to broadcast at most every 200ms instead of on every audio chunk, reducing frame processing overhead during user speech.
diff --git a/changelog/3484.fixed.md b/changelog/3484.fixed.md
deleted file mode 100644
index 9f2823a3e..000000000
--- a/changelog/3484.fixed.md
+++ /dev/null
@@ -1 +0,0 @@
-- Fixed a `Mem0MemoryService` issue where passing `async_mode: true` was causing an error. See https://docs.mem0.ai/platform/features/async-mode-default-change.
diff --git a/changelog/3486.fixed.md b/changelog/3486.fixed.md
deleted file mode 100644
index a02427e6e..000000000
--- a/changelog/3486.fixed.md
+++ /dev/null
@@ -1 +0,0 @@
-- Fixed `AWSNovaSonicLLMService.reset_conversation()`, which would previously error out. Now it successfully reconnects and "rehydrates" from the context object.
diff --git a/changelog/3489.fixed.md b/changelog/3489.fixed.md
deleted file mode 100644
index c61b25444..000000000
--- a/changelog/3489.fixed.md
+++ /dev/null
@@ -1,3 +0,0 @@
-- Fixed `AzureTTSService` transcript formatting issues:
-  - Punctuation now appears without extra spaces (e.g., "Hello!" instead of "Hello !")
-  - CJK languages (Chinese, Japanese, Korean) no longer have unwanted spaces between characters
diff --git a/changelog/3490.added.md b/changelog/3490.added.md
deleted file mode 100644
index 905d35b34..000000000
--- a/changelog/3490.added.md
+++ /dev/null
@@ -1 +0,0 @@
-- Added `on_user_mute_started` and `on_user_mute_stopped` event handlers to `LLMUserAggregator` for tracking user mute state changes.
diff --git a/changelog/3494.fixed.md b/changelog/3494.fixed.md
deleted file mode 100644
index 41fbfa74e..000000000
--- a/changelog/3494.fixed.md
+++ /dev/null
@@ -1 +0,0 @@
-- Fixed an issue where `UninterruptibleFrame` frames would not be preserved in some cases.
diff --git a/changelog/3499.fixed.md b/changelog/3499.fixed.md
deleted file mode 100644
index ae893c13d..000000000
--- a/changelog/3499.fixed.md
+++ /dev/null
@@ -1 +0,0 @@
-- Fixed memory leak in `LiveKitTransport` when `video_in_enabled` is `False`.
diff --git a/changelog/3503.fixed.md b/changelog/3503.fixed.md
deleted file mode 100644
index 4218f8781..000000000
--- a/changelog/3503.fixed.md
+++ /dev/null
@@ -1 +0,0 @@
-- Fixed an issue in `AIService` where unhandled exceptions in `start()`, `stop()`, or `cancel()` implementations would prevent `process_frame()` to continue and therefore `StartFrame`, `EndFrame`, or `CancelFrame` from being pushed downstream, causing the pipeline to not start or stop properly.
diff --git a/changelog/3504.fixed.md b/changelog/3504.fixed.md
deleted file mode 100644
index 2d675547a..000000000
--- a/changelog/3504.fixed.md
+++ /dev/null
@@ -1 +0,0 @@
-- Moved `NVIDIATTSService` and `NVIDIASTTService` client initialization from constructor to `start()` for better error handling.
diff --git a/changelog/3509.fixed.2.md b/changelog/3509.fixed.2.md
deleted file mode 100644
index 68053011b..000000000
--- a/changelog/3509.fixed.2.md
+++ /dev/null
@@ -1 +0,0 @@
-- Optimized `NVIDIATTSService` to process incoming audio frames immediately.
diff --git a/changelog/3509.fixed.md b/changelog/3509.fixed.md
deleted file mode 100644
index 153060b46..000000000
--- a/changelog/3509.fixed.md
+++ /dev/null
@@ -1 +0,0 @@
-- Optimized `NVIDIASTTService` by removing unnecessary queue and task.
diff --git a/changelog/3511.fixed.md b/changelog/3511.fixed.md
deleted file mode 100644
index 1f4f429ac..000000000
--- a/changelog/3511.fixed.md
+++ /dev/null
@@ -1 +0,0 @@
-- Fixed a `CambTTSService` issue where client was being initialized in the constructor which wouldn't allow for proper Pipeline error handling.

From 4a724379fcd59aa20e2b203bc8df9a8c3eca8ae7 Mon Sep 17 00:00:00 2001
From: okue <nogideca@gmail.com>
Date: Wed, 21 Jan 2026 23:58:49 +0900
Subject: [PATCH 25/51] refactor(user_mute): remove unnecessary _bot_speaking
 assignment in _handle_bot_stopped_speaking

The _bot_speaking flag does not need to be set in this method,
so the redundant assignment has been removed.
---
 .../mute_until_first_bot_complete_user_mute_strategy.py          | 1 -
 1 file changed, 1 deletion(-)

diff --git a/src/pipecat/turns/user_mute/mute_until_first_bot_complete_user_mute_strategy.py b/src/pipecat/turns/user_mute/mute_until_first_bot_complete_user_mute_strategy.py
index 20fe7016a..c19499e07 100644
--- a/src/pipecat/turns/user_mute/mute_until_first_bot_complete_user_mute_strategy.py
+++ b/src/pipecat/turns/user_mute/mute_until_first_bot_complete_user_mute_strategy.py
@@ -51,6 +51,5 @@ class MuteUntilFirstBotCompleteUserMuteStrategy(BaseUserMuteStrategy):
         return not self._first_speech_handled
 
     async def _handle_bot_stopped_speaking(self, frame: BotStoppedSpeakingFrame):
-        self._bot_speaking = False
         if not self._first_speech_handled:
             self._first_speech_handled = True

From eacd2a4b71071dbd9108afa838ba637b0c39c350 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Aleix=20Conchillo=20Flaqu=C3=A9?= <aleix@daily.co>
Date: Wed, 21 Jan 2026 14:10:56 -0800
Subject: [PATCH 26/51] FrameProcessor: add broadcast_frame_instance()

---
 src/pipecat/processors/frame_processor.py |  34 +++++
 src/pipecat/serializers/protobuf.py       |   2 +-
 tests/test_frame_processor.py             | 166 +++++++++++++++++++++-
 3 files changed, 200 insertions(+), 2 deletions(-)

diff --git a/src/pipecat/processors/frame_processor.py b/src/pipecat/processors/frame_processor.py
index 70f44dfca..10aafc5a8 100644
--- a/src/pipecat/processors/frame_processor.py
+++ b/src/pipecat/processors/frame_processor.py
@@ -12,7 +12,9 @@ management, and frame flow control mechanisms.
 """
 
 import asyncio
+import dataclasses
 import traceback
+from copy import deepcopy
 from dataclasses import dataclass
 from enum import Enum
 from typing import (
@@ -782,6 +784,38 @@ class FrameProcessor(BaseObject):
         await self.push_frame(frame_cls(**kwargs))
         await self.push_frame(frame_cls(**kwargs), FrameDirection.UPSTREAM)
 
+    async def broadcast_frame_instance(self, frame: Frame):
+        """Broadcasts a frame instance upstream and downstream.
+
+        This method extracts the class and init fields from the given frame
+        instance and creates two new instances to push upstream and downstream.
+
+        Args:
+            frame: The frame instance to broadcast.
+
+        Note:
+            Prefer using `broadcast_frame()` when possible, as it is more
+            efficient. This method should only be used when you are not the
+            creator of the frame and need to broadcast an existing instance.
+        """
+        frame_cls = type(frame)
+        init_fields = {f.name: getattr(frame, f.name) for f in dataclasses.fields(frame) if f.init}
+        extra_fields = {
+            f.name: getattr(frame, f.name)
+            for f in dataclasses.fields(frame)
+            if not f.init and f.name not in ("id", "name")
+        }
+
+        new_frame = frame_cls(**deepcopy(init_fields))
+        for k, v in deepcopy(extra_fields).items():
+            setattr(new_frame, k, v)
+        await self.push_frame(new_frame)
+
+        new_frame = frame_cls(**deepcopy(init_fields))
+        for k, v in deepcopy(extra_fields).items():
+            setattr(new_frame, k, v)
+        await self.push_frame(new_frame, FrameDirection.UPSTREAM)
+
     async def __start(self, frame: StartFrame):
         """Handle the start frame to initialize processor state.
 
diff --git a/src/pipecat/serializers/protobuf.py b/src/pipecat/serializers/protobuf.py
index 6d989c7dd..912f20b42 100644
--- a/src/pipecat/serializers/protobuf.py
+++ b/src/pipecat/serializers/protobuf.py
@@ -126,7 +126,7 @@ class ProtobufFrameSerializer(FrameSerializer):
         if "pts" in args_dict:
             del args_dict["pts"]
 
-        # Special handling for MessageFrame -> OutputTransportMessageUrgentFrame
+        # Special handling for MessageFrame -> InputTransportMessageFrame
         if class_name == MessageFrame:
             try:
                 msg = json.loads(args_dict["data"])
diff --git a/tests/test_frame_processor.py b/tests/test_frame_processor.py
index b3a11ab43..3f8c8ae70 100644
--- a/tests/test_frame_processor.py
+++ b/tests/test_frame_processor.py
@@ -6,7 +6,8 @@
 
 import asyncio
 import unittest
-from dataclasses import dataclass
+from dataclasses import dataclass, field
+from typing import List
 
 from pipecat.frames.frames import (
     DataFrame,
@@ -24,6 +25,15 @@ from pipecat.processors.frame_processor import FrameDirection, FrameProcessor
 from pipecat.tests.utils import SleepFrame, run_test
 
 
+@dataclass
+class BroadcastTestFrame(DataFrame):
+    """Test frame with init fields for broadcast testing."""
+
+    text: str = ""
+    value: int = 0
+    items: List[str] = field(default_factory=list)
+
+
 class TestFrameProcessor(unittest.IsolatedAsyncioTestCase):
     async def test_before_after_events(self):
         identity = IdentityFilter()
@@ -186,3 +196,157 @@ class TestFrameProcessor(unittest.IsolatedAsyncioTestCase):
             frames_to_send=frames_to_send,
             expected_down_frames=expected_down_frames,
         )
+
+    async def test_broadcast_frame(self):
+        """Test that broadcast_frame creates two separate frames with fresh IDs."""
+        downstream_frames: List[Frame] = []
+        upstream_frames: List[Frame] = []
+
+        class BroadcastTestProcessor(FrameProcessor):
+            async def process_frame(self, frame: Frame, direction: FrameDirection):
+                await super().process_frame(frame, direction)
+                if isinstance(frame, TextFrame):
+                    await self.broadcast_frame(
+                        BroadcastTestFrame, text="hello", value=42, items=["a", "b"]
+                    )
+                else:
+                    await self.push_frame(frame, direction)
+
+        class CaptureProcessor(FrameProcessor):
+            def __init__(self, capture_list: List[Frame], direction: FrameDirection):
+                super().__init__()
+                self._capture_list = capture_list
+                self._capture_direction = direction
+
+            async def process_frame(self, frame: Frame, direction: FrameDirection):
+                await super().process_frame(frame, direction)
+                if direction == self._capture_direction and isinstance(frame, BroadcastTestFrame):
+                    self._capture_list.append(frame)
+                await self.push_frame(frame, direction)
+
+        up_capture = CaptureProcessor(upstream_frames, FrameDirection.UPSTREAM)
+        broadcaster = BroadcastTestProcessor()
+        down_capture = CaptureProcessor(downstream_frames, FrameDirection.DOWNSTREAM)
+
+        pipeline = Pipeline([up_capture, broadcaster, down_capture])
+
+        frames_to_send = [TextFrame(text="trigger")]
+        expected_down_frames = [BroadcastTestFrame]
+        expected_up_frames = [BroadcastTestFrame]
+
+        await run_test(
+            pipeline,
+            frames_to_send=frames_to_send,
+            expected_down_frames=expected_down_frames,
+            expected_up_frames=expected_up_frames,
+        )
+
+        # Verify we got one frame in each direction
+        self.assertEqual(len(downstream_frames), 1)
+        self.assertEqual(len(upstream_frames), 1)
+
+        down_frame = downstream_frames[0]
+        up_frame = upstream_frames[0]
+
+        # Verify the frames have different IDs (they are separate instances)
+        self.assertNotEqual(down_frame.id, up_frame.id)
+
+        # Verify the frames have the correct field values
+        self.assertEqual(down_frame.text, "hello")
+        self.assertEqual(down_frame.value, 42)
+        self.assertEqual(down_frame.items, ["a", "b"])
+        self.assertEqual(up_frame.text, "hello")
+        self.assertEqual(up_frame.value, 42)
+        self.assertEqual(up_frame.items, ["a", "b"])
+
+        # Verify the items lists are separate instances (not shared references)
+        self.assertIsNot(down_frame.items, up_frame.items)
+
+    async def test_broadcast_frame_instance(self):
+        """Test that broadcast_frame_instance copies all fields except id and name."""
+        downstream_frames: List[Frame] = []
+        upstream_frames: List[Frame] = []
+        original_frame: List[Frame] = []
+
+        class BroadcastInstanceTestProcessor(FrameProcessor):
+            async def process_frame(self, frame: Frame, direction: FrameDirection):
+                await super().process_frame(frame, direction)
+                if isinstance(frame, BroadcastTestFrame):
+                    # Set some non-init fields on the frame
+                    frame.pts = 12345
+                    frame.metadata = {"key": "value", "nested": {"a": 1}}
+                    original_frame.append(frame)
+                    await self.broadcast_frame_instance(frame)
+                else:
+                    await self.push_frame(frame, direction)
+
+        class CaptureProcessor(FrameProcessor):
+            def __init__(self, capture_list: List[Frame], direction: FrameDirection):
+                super().__init__()
+                self._capture_list = capture_list
+                self._capture_direction = direction
+
+            async def process_frame(self, frame: Frame, direction: FrameDirection):
+                await super().process_frame(frame, direction)
+                if direction == self._capture_direction and isinstance(frame, BroadcastTestFrame):
+                    self._capture_list.append(frame)
+                await self.push_frame(frame, direction)
+
+        up_capture = CaptureProcessor(upstream_frames, FrameDirection.UPSTREAM)
+        broadcaster = BroadcastInstanceTestProcessor()
+        down_capture = CaptureProcessor(downstream_frames, FrameDirection.DOWNSTREAM)
+
+        pipeline = Pipeline([up_capture, broadcaster, down_capture])
+
+        # Create a frame with mutable fields to test deep copying
+        test_frame = BroadcastTestFrame(text="test", value=99, items=["x", "y", "z"])
+
+        frames_to_send = [test_frame]
+        expected_down_frames = [BroadcastTestFrame]
+        expected_up_frames = [BroadcastTestFrame]
+
+        await run_test(
+            pipeline,
+            frames_to_send=frames_to_send,
+            expected_down_frames=expected_down_frames,
+            expected_up_frames=expected_up_frames,
+        )
+
+        # Verify we got one frame in each direction
+        self.assertEqual(len(downstream_frames), 1)
+        self.assertEqual(len(upstream_frames), 1)
+        self.assertEqual(len(original_frame), 1)
+
+        orig = original_frame[0]
+        down_frame = downstream_frames[0]
+        up_frame = upstream_frames[0]
+
+        # Verify the frames have different IDs and names (fresh values)
+        self.assertNotEqual(down_frame.id, orig.id)
+        self.assertNotEqual(up_frame.id, orig.id)
+        self.assertNotEqual(down_frame.id, up_frame.id)
+        self.assertNotEqual(down_frame.name, orig.name)
+        self.assertNotEqual(up_frame.name, orig.name)
+
+        # Verify init fields are copied correctly
+        self.assertEqual(down_frame.text, "test")
+        self.assertEqual(down_frame.value, 99)
+        self.assertEqual(down_frame.items, ["x", "y", "z"])
+        self.assertEqual(up_frame.text, "test")
+        self.assertEqual(up_frame.value, 99)
+        self.assertEqual(up_frame.items, ["x", "y", "z"])
+
+        # Verify non-init fields (except id/name) are copied
+        self.assertEqual(down_frame.pts, 12345)
+        self.assertEqual(down_frame.metadata, {"key": "value", "nested": {"a": 1}})
+        self.assertEqual(up_frame.pts, 12345)
+        self.assertEqual(up_frame.metadata, {"key": "value", "nested": {"a": 1}})
+
+        # Verify mutable fields are deep copied (not shared references)
+        self.assertIsNot(down_frame.items, orig.items)
+        self.assertIsNot(up_frame.items, orig.items)
+        self.assertIsNot(down_frame.items, up_frame.items)
+        self.assertIsNot(down_frame.metadata, orig.metadata)
+        self.assertIsNot(up_frame.metadata, orig.metadata)
+        self.assertIsNot(down_frame.metadata, up_frame.metadata)
+        self.assertIsNot(down_frame.metadata["nested"], up_frame.metadata["nested"])

From ba0ddb1832a62297c214e4f60fcf5235cf933a04 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Aleix=20Conchillo=20Flaqu=C3=A9?= <aleix@daily.co>
Date: Wed, 21 Jan 2026 18:05:15 -0800
Subject: [PATCH 27/51] FrameProcessor: copy kwargs when broadcasting frame

---
 src/pipecat/processors/frame_processor.py | 8 ++++----
 1 file changed, 4 insertions(+), 4 deletions(-)

diff --git a/src/pipecat/processors/frame_processor.py b/src/pipecat/processors/frame_processor.py
index 10aafc5a8..1ad57e0d6 100644
--- a/src/pipecat/processors/frame_processor.py
+++ b/src/pipecat/processors/frame_processor.py
@@ -781,14 +781,14 @@ class FrameProcessor(BaseObject):
             frame_cls: The class of the frame to be broadcasted.
             **kwargs: Keyword arguments to be passed to the frame's constructor.
         """
-        await self.push_frame(frame_cls(**kwargs))
-        await self.push_frame(frame_cls(**kwargs), FrameDirection.UPSTREAM)
+        await self.push_frame(frame_cls(**deepcopy(kwargs)))
+        await self.push_frame(frame_cls(**deepcopy(kwargs)), FrameDirection.UPSTREAM)
 
     async def broadcast_frame_instance(self, frame: Frame):
         """Broadcasts a frame instance upstream and downstream.
 
-        This method extracts the class and init fields from the given frame
-        instance and creates two new instances to push upstream and downstream.
+        This method creates two new frame instances copying all fields from the
+        original frame except `id` and `name`, which get fresh values.
 
         Args:
             frame: The frame instance to broadcast.

From 62f4708d43a4f393ff386e305ed37304ed48ab57 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Aleix=20Conchillo=20Flaqu=C3=A9?= <aleix@daily.co>
Date: Wed, 21 Jan 2026 14:12:47 -0800
Subject: [PATCH 28/51] transports: broadcast InputTransportMessageFrame frames

---
 src/pipecat/transports/daily/transport.py       | 5 +++--
 src/pipecat/transports/smallwebrtc/transport.py | 3 +--
 src/pipecat/transports/websocket/client.py      | 3 +++
 src/pipecat/transports/websocket/fastapi.py     | 3 +++
 src/pipecat/transports/websocket/server.py      | 4 ++++
 5 files changed, 14 insertions(+), 4 deletions(-)

diff --git a/src/pipecat/transports/daily/transport.py b/src/pipecat/transports/daily/transport.py
index 3b652b517..e220ab0d8 100644
--- a/src/pipecat/transports/daily/transport.py
+++ b/src/pipecat/transports/daily/transport.py
@@ -1728,8 +1728,9 @@ class DailyInputTransport(BaseInputTransport):
             message: The message data to send.
             sender: ID of the message sender.
         """
-        frame = DailyInputTransportMessageFrame(message=message, participant_id=sender)
-        await self.push_frame(frame)
+        await self.broadcast_frame_class(
+            DailyInputTransportMessageFrame, message=message, participant_id=sender
+        )
 
     #
     # Audio in
diff --git a/src/pipecat/transports/smallwebrtc/transport.py b/src/pipecat/transports/smallwebrtc/transport.py
index e9a07cf40..bcc3c8b79 100644
--- a/src/pipecat/transports/smallwebrtc/transport.py
+++ b/src/pipecat/transports/smallwebrtc/transport.py
@@ -698,8 +698,7 @@ class SmallWebRTCInputTransport(BaseInputTransport):
             message: The application message to process.
         """
         logger.debug(f"Received app message inside SmallWebRTCInputTransport  {message}")
-        frame = InputTransportMessageFrame(message=message)
-        await self.push_frame(frame)
+        await self.broadcast_frame_class(InputTransportMessageFrame, message=message)
 
     # Add this method similar to DailyInputTransport.request_participant_image
     async def request_participant_image(self, frame: UserImageRequestFrame):
diff --git a/src/pipecat/transports/websocket/client.py b/src/pipecat/transports/websocket/client.py
index 02c891c54..f55b1f919 100644
--- a/src/pipecat/transports/websocket/client.py
+++ b/src/pipecat/transports/websocket/client.py
@@ -27,6 +27,7 @@ from pipecat.frames.frames import (
     EndFrame,
     Frame,
     InputAudioRawFrame,
+    InputTransportMessageFrame,
     OutputAudioRawFrame,
     OutputTransportMessageFrame,
     OutputTransportMessageUrgentFrame,
@@ -298,6 +299,8 @@ class WebsocketClientInputTransport(BaseInputTransport):
             return
         if isinstance(frame, InputAudioRawFrame) and self._params.audio_in_enabled:
             await self.push_audio_frame(frame)
+        elif isinstance(frame, InputTransportMessageFrame):
+            await self.broadcast_frame(frame)
         else:
             await self.push_frame(frame)
 
diff --git a/src/pipecat/transports/websocket/fastapi.py b/src/pipecat/transports/websocket/fastapi.py
index e1d02ac00..a0d02ccec 100644
--- a/src/pipecat/transports/websocket/fastapi.py
+++ b/src/pipecat/transports/websocket/fastapi.py
@@ -26,6 +26,7 @@ from pipecat.frames.frames import (
     EndFrame,
     Frame,
     InputAudioRawFrame,
+    InputTransportMessageFrame,
     InterruptionFrame,
     OutputAudioRawFrame,
     OutputTransportMessageFrame,
@@ -311,6 +312,8 @@ class FastAPIWebsocketInputTransport(BaseInputTransport):
 
                 if isinstance(frame, InputAudioRawFrame):
                     await self.push_audio_frame(frame)
+                elif isinstance(frame, InputTransportMessageFrame):
+                    await self.broadcast_frame(frame)
                 else:
                     await self.push_frame(frame)
         except Exception as e:
diff --git a/src/pipecat/transports/websocket/server.py b/src/pipecat/transports/websocket/server.py
index a31ac5487..acd4faf92 100644
--- a/src/pipecat/transports/websocket/server.py
+++ b/src/pipecat/transports/websocket/server.py
@@ -25,6 +25,8 @@ from pipecat.frames.frames import (
     EndFrame,
     Frame,
     InputAudioRawFrame,
+    InputTransportMessageFrame,
+    InputTransportMessageUrgentFrame,
     InterruptionFrame,
     OutputAudioRawFrame,
     OutputTransportMessageFrame,
@@ -214,6 +216,8 @@ class WebsocketServerInputTransport(BaseInputTransport):
 
                 if isinstance(frame, InputAudioRawFrame):
                     await self.push_audio_frame(frame)
+                elif isinstance(frame, InputTransportMessageFrame):
+                    await self.broadcast_frame(frame)
                 else:
                     await self.push_frame(frame)
         except Exception as e:

From cc61cdbba37ad191e0049de463485b7ccdb09b56 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Aleix=20Conchillo=20Flaqu=C3=A9?= <aleix@daily.co>
Date: Wed, 21 Jan 2026 14:13:17 -0800
Subject: [PATCH 29/51] RTVIProcessor: add create_rtvi_observer()

---
 src/pipecat/processors/frameworks/rtvi.py | 12 +++++++++++
 src/pipecat/services/google/rtvi.py       | 25 ++++++++++++++++++++---
 2 files changed, 34 insertions(+), 3 deletions(-)

diff --git a/src/pipecat/processors/frameworks/rtvi.py b/src/pipecat/processors/frameworks/rtvi.py
index 7ea04ecf2..ee25af839 100644
--- a/src/pipecat/processors/frameworks/rtvi.py
+++ b/src/pipecat/processors/frameworks/rtvi.py
@@ -1413,6 +1413,18 @@ class RTVIProcessor(FrameProcessor):
 
         self._registered_services[service.name] = service
 
+    def create_rtvi_observer(self, *, params: Optional[RTVIObserverParams] = None, **kwargs):
+        """Creates a new RTVI Observer.
+
+        Args:
+            params: Settings to enable/disable specific messages.
+            **kwargs: Additional arguments passed to the observer.
+
+        Returns:
+            A new RTVI observer.
+        """
+        return RTVIObserver(self, params=params, **kwargs)
+
     async def set_client_ready(self):
         """Mark the client as ready and trigger the ready event."""
         self._client_ready = True
diff --git a/src/pipecat/services/google/rtvi.py b/src/pipecat/services/google/rtvi.py
index 1ef70d67d..738b0ab9d 100644
--- a/src/pipecat/services/google/rtvi.py
+++ b/src/pipecat/services/google/rtvi.py
@@ -4,7 +4,7 @@
 # SPDX-License-Identifier: BSD 2-Clause License
 #
 
-"""Google RTVI integration models and observer implementation.
+"""Google RTVI processor and observer implementation.
 
 This module provides integration with Google's services through the RTVI framework,
 including models for search responses and an observer for handling Google-specific
@@ -16,7 +16,7 @@ from typing import List, Literal, Optional
 from pydantic import BaseModel
 
 from pipecat.observers.base_observer import FramePushed
-from pipecat.processors.frameworks.rtvi import RTVIObserver, RTVIProcessor
+from pipecat.processors.frameworks.rtvi import RTVIObserver, RTVIObserverParams, RTVIProcessor
 from pipecat.services.google.frames import LLMSearchOrigin, LLMSearchResponseFrame
 
 
@@ -86,4 +86,23 @@ class GoogleRTVIObserver(RTVIObserver):
                 rendered_content=frame.rendered_content,
             )
         )
-        await self.push_transport_message_urgent(message)
+        await self.send_rtvi_message(message)
+
+
+class GoogleRTVIProcessor(RTVIProcessor):
+    """RTVI processor for Google service integration.
+
+    Creates a specific Google RTVI Observer.
+    """
+
+    def create_rtvi_observer(self, *, params: Optional[RTVIObserverParams] = None, **kwargs):
+        """Creates a new RTVI Observer.
+
+        Args:
+            params: Settings to enable/disable specific messages.
+            **kwargs: Additional arguments passed to the observer.
+
+        Returns:
+            A new RTVI observer.
+        """
+        return GoogleRTVIObserver(self)

From e85a00cc0ee435eff8eb931c11aa0bccd018bed4 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Aleix=20Conchillo=20Flaqu=C3=A9?= <aleix@daily.co>
Date: Wed, 21 Jan 2026 14:14:43 -0800
Subject: [PATCH 30/51] PipelineTask: automatically add RTVI processor and RTVI
 observer

If `enable_rtvi` is enabled (enabled by default) and RTVI processor will be
added automatically to the pipeline. Also, and RTVI observer will be
registered.
---
 src/pipecat/pipeline/task.py | 31 ++++++++++++++++++++++++++++++-
 1 file changed, 30 insertions(+), 1 deletion(-)

diff --git a/src/pipecat/pipeline/task.py b/src/pipecat/pipeline/task.py
index 6d4b4c039..9a2f1d33d 100644
--- a/src/pipecat/pipeline/task.py
+++ b/src/pipecat/pipeline/task.py
@@ -49,6 +49,7 @@ from pipecat.pipeline.pipeline import Pipeline, PipelineSink, PipelineSource
 from pipecat.pipeline.task_observer import TaskObserver
 from pipecat.processors.aggregators.llm_response import LLMUserContextAggregator
 from pipecat.processors.frame_processor import FrameDirection, FrameProcessor, FrameProcessorSetup
+from pipecat.processors.frameworks.rtvi import RTVIObserverParams, RTVIProcessor
 from pipecat.utils.asyncio.task_manager import BaseTaskManager, TaskManager, TaskManagerParams
 from pipecat.utils.tracing.setup import is_tracing_available
 from pipecat.utils.tracing.turn_trace_observer import TurnTraceObserver
@@ -225,9 +226,12 @@ class PipelineTask(BasePipelineTask):
         conversation_id: Optional[str] = None,
         enable_tracing: bool = False,
         enable_turn_tracking: bool = True,
+        enable_rtvi: bool = True,
         idle_timeout_frames: Tuple[Type[Frame], ...] = (BotSpeakingFrame, UserSpeakingFrame),
         idle_timeout_secs: Optional[float] = IDLE_TIMEOUT_SECS,
         observers: Optional[List[BaseObserver]] = None,
+        rtvi_processor: Optional[RTVIProcessor] = None,
+        rtvi_observer_params: Optional[RTVIObserverParams] = None,
         task_manager: Optional[BaseTaskManager] = None,
     ):
         """Initialize the PipelineTask.
@@ -244,6 +248,7 @@ class PipelineTask(BasePipelineTask):
             check_dangling_tasks: Whether to check for processors' tasks finishing properly.
             clock: Clock implementation for timing operations.
             conversation_id: Optional custom ID for the conversation.
+            enable_rtvi: Whether to automatically add RTVI support to the pipeline.
             enable_tracing: Whether to enable tracing.
             enable_turn_tracking: Whether to enable turn tracking.
             idle_timeout_frames: A tuple with the frames that should trigger an idle
@@ -252,6 +257,8 @@ class PipelineTask(BasePipelineTask):
                 None. If a pipeline is idle the pipeline task will be cancelled
                 automatically.
             observers: List of observers for monitoring pipeline execution.
+            rtvi_observer_params: The RTVI observer parameter to use if RTVI is enabled.
+            rtvi_processor: The RTVI processor to add if RTVI is enabled.
             task_manager: Optional task manager for handling asyncio tasks.
         """
         super().__init__()
@@ -306,6 +313,16 @@ class PipelineTask(BasePipelineTask):
         self._heartbeat_push_task: Optional[asyncio.Task] = None
         self._heartbeat_monitor_task: Optional[asyncio.Task] = None
 
+        # RTVI support
+        self._rtvi = None
+        if enable_rtvi:
+            self._rtvi = rtvi_processor or RTVIProcessor()
+            observers.append(self._rtvi.create_rtvi_observer(params=rtvi_observer_params))
+
+            @self.rtvi.event_handler("on_client_ready")
+            async def on_client_ready(rtvi: RTVIProcessor):
+                await rtvi.set_bot_ready()
+
         # This is the idle event. When selected frames are pushed from any
         # processor we consider the pipeline is not idle. We use an observer
         # which will be listening any part of the pipeline.
@@ -335,7 +352,8 @@ class PipelineTask(BasePipelineTask):
         # allows us to receive and react to downstream frames.
         source = PipelineSource(self._source_push_frame, name=f"{self}::Source")
         sink = PipelineSink(self._sink_push_frame, name=f"{self}::Sink")
-        self._pipeline = Pipeline([pipeline], source=source, sink=sink)
+        processors = [self._rtvi, pipeline] if self._rtvi else [pipeline]
+        self._pipeline = Pipeline(processors, source=source, sink=sink)
 
         # The task observer acts as a proxy to the provided observers. This way,
         # we only need to pass a single observer (using the StartFrame) which
@@ -398,6 +416,17 @@ class PipelineTask(BasePipelineTask):
         """
         return self._turn_trace_observer
 
+    @property
+    def rtvi(self) -> RTVIProcessor:
+        """Get the RTVI processor if RTVI is enabled.
+
+        Returns:
+            The RTVI processor added to the pipeline when RTVI is enabled.
+        """
+        if not self._rtvi:
+            raise Exception(f"{self} RTVI is not enabled.")
+        return self._rtvi
+
     def event_handler(self, event_name: str):
         """Decorator for registering event handlers.
 

From 054e50486824a9b18a3c0faa0556d44cfb940d0f Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Aleix=20Conchillo=20Flaqu=C3=A9?= <aleix@daily.co>
Date: Wed, 21 Jan 2026 14:16:51 -0800
Subject: [PATCH 31/51] examples(foundational): remove RTVI (automatically
 added by PipelineTask)

---
 .../foundational/07zb-interruptible-inworld-http.py   |  5 -----
 examples/foundational/07zb-interruptible-inworld.py   |  5 -----
 examples/foundational/07ze-interruptible-hume.py      |  9 ---------
 examples/foundational/37-mem0.py                      |  7 +------
 examples/foundational/38b-smart-turn-local.py         |  7 +------
 examples/foundational/46-video-processing.py          | 11 ++---------
 6 files changed, 4 insertions(+), 40 deletions(-)

diff --git a/examples/foundational/07zb-interruptible-inworld-http.py b/examples/foundational/07zb-interruptible-inworld-http.py
index b2bf05660..c67984f7c 100644
--- a/examples/foundational/07zb-interruptible-inworld-http.py
+++ b/examples/foundational/07zb-interruptible-inworld-http.py
@@ -23,7 +23,6 @@ from pipecat.processors.aggregators.llm_response_universal import (
     LLMContextAggregatorPair,
     LLMUserAggregatorParams,
 )
-from pipecat.processors.frameworks.rtvi import RTVIObserver, RTVIProcessor
 from pipecat.runner.types import RunnerArguments
 from pipecat.runner.utils import create_transport
 from pipecat.services.deepgram.stt import DeepgramSTTService
@@ -93,12 +92,9 @@ async def run_bot(transport: BaseTransport, runner_args: RunnerArguments):
             ),
         )
 
-        rtvi = RTVIProcessor()
-
         pipeline = Pipeline(
             [
                 transport.input(),
-                rtvi,
                 stt,
                 user_aggregator,
                 llm,
@@ -115,7 +111,6 @@ async def run_bot(transport: BaseTransport, runner_args: RunnerArguments):
                 enable_usage_metrics=True,
             ),
             observers=[
-                RTVIObserver(rtvi),
                 DebugLogObserver(
                     frame_types={
                         TTSTextFrame: (BaseOutputTransport, FrameEndpoint.SOURCE),
diff --git a/examples/foundational/07zb-interruptible-inworld.py b/examples/foundational/07zb-interruptible-inworld.py
index 8d3d351a5..46e07dc5f 100644
--- a/examples/foundational/07zb-interruptible-inworld.py
+++ b/examples/foundational/07zb-interruptible-inworld.py
@@ -22,7 +22,6 @@ from pipecat.processors.aggregators.llm_response_universal import (
     LLMContextAggregatorPair,
     LLMUserAggregatorParams,
 )
-from pipecat.processors.frameworks.rtvi import RTVIConfig, RTVIObserver, RTVIProcessor
 from pipecat.runner.types import RunnerArguments
 from pipecat.runner.utils import create_transport
 from pipecat.services.deepgram.stt import DeepgramSTTService
@@ -88,12 +87,9 @@ async def run_bot(transport: BaseTransport, runner_args: RunnerArguments):
         ),
     )
 
-    rtvi = RTVIProcessor(config=RTVIConfig(config=[]))
-
     pipeline = Pipeline(
         [
             transport.input(),
-            rtvi,
             stt,
             user_aggregator,
             llm,
@@ -110,7 +106,6 @@ async def run_bot(transport: BaseTransport, runner_args: RunnerArguments):
             enable_usage_metrics=True,
         ),
         observers=[
-            RTVIObserver(rtvi),
             DebugLogObserver(
                 frame_types={
                     TTSTextFrame: (BaseOutputTransport, FrameEndpoint.SOURCE),
diff --git a/examples/foundational/07ze-interruptible-hume.py b/examples/foundational/07ze-interruptible-hume.py
index 33ea4d278..e9a8e355c 100644
--- a/examples/foundational/07ze-interruptible-hume.py
+++ b/examples/foundational/07ze-interruptible-hume.py
@@ -22,7 +22,6 @@ from pipecat.processors.aggregators.llm_response_universal import (
     LLMContextAggregatorPair,
     LLMUserAggregatorParams,
 )
-from pipecat.processors.frameworks.rtvi import RTVIConfig, RTVIObserver, RTVIProcessor
 from pipecat.runner.types import RunnerArguments
 from pipecat.runner.utils import create_transport
 from pipecat.services.deepgram.stt import DeepgramSTTService
@@ -90,12 +89,9 @@ async def run_bot(transport: BaseTransport, runner_args: RunnerArguments):
         ),
     )
 
-    rtvi = RTVIProcessor(config=RTVIConfig(config=[]))
-
     pipeline = Pipeline(
         [
             transport.input(),  # Transport user input
-            rtvi,
             stt,
             user_aggregator,  # User responses
             llm,  # LLM
@@ -114,7 +110,6 @@ async def run_bot(transport: BaseTransport, runner_args: RunnerArguments):
         ),
         idle_timeout_secs=runner_args.pipeline_idle_timeout_secs,
         observers=[
-            RTVIObserver(rtvi),
             DebugLogObserver(
                 frame_types={
                     TTSTextFrame: (BaseOutputTransport, FrameEndpoint.SOURCE),
@@ -123,10 +118,6 @@ async def run_bot(transport: BaseTransport, runner_args: RunnerArguments):
         ],
     )
 
-    @rtvi.event_handler("on_client_ready")
-    async def on_client_ready(rtvi):
-        await rtvi.set_bot_ready()
-
     @transport.event_handler("on_client_connected")
     async def on_client_connected(transport, client):
         logger.info(f"Client connected")
diff --git a/examples/foundational/37-mem0.py b/examples/foundational/37-mem0.py
index c82d39edc..08a739c0a 100644
--- a/examples/foundational/37-mem0.py
+++ b/examples/foundational/37-mem0.py
@@ -59,7 +59,6 @@ from pipecat.processors.aggregators.llm_response_universal import (
     LLMContextAggregatorPair,
     LLMUserAggregatorParams,
 )
-from pipecat.processors.frameworks.rtvi import RTVIConfig, RTVIObserver, RTVIProcessor
 from pipecat.runner.types import RunnerArguments
 from pipecat.runner.utils import create_transport
 from pipecat.services.deepgram.stt import DeepgramSTTService
@@ -255,12 +254,10 @@ async def run_bot(transport: BaseTransport, runner_args: RunnerArguments):
             ),
         ),
     )
-    rtvi = RTVIProcessor(config=RTVIConfig(config=[]))
 
     pipeline = Pipeline(
         [
             transport.input(),
-            rtvi,
             stt,
             user_aggregator,
             memory,
@@ -278,12 +275,10 @@ async def run_bot(transport: BaseTransport, runner_args: RunnerArguments):
             enable_usage_metrics=True,
         ),
         idle_timeout_secs=runner_args.pipeline_idle_timeout_secs,
-        observers=[RTVIObserver(rtvi)],
     )
 
-    @rtvi.event_handler("on_client_ready")
+    @task.rtvi.event_handler("on_client_ready")
     async def on_client_ready(rtvi):
-        await rtvi.set_bot_ready()
         # Get personalized greeting based on user memories. Can pass agent_id and run_id as per requirement of the application to manage short term memory or agent specific memory.
         greeting = await get_initial_greeting(
             memory_client=memory.memory_client, user_id=USER_ID, agent_id=None, run_id=None
diff --git a/examples/foundational/38b-smart-turn-local.py b/examples/foundational/38b-smart-turn-local.py
index 37112de4d..85a320631 100644
--- a/examples/foundational/38b-smart-turn-local.py
+++ b/examples/foundational/38b-smart-turn-local.py
@@ -22,7 +22,6 @@ from pipecat.processors.aggregators.llm_response_universal import (
     LLMContextAggregatorPair,
     LLMUserAggregatorParams,
 )
-from pipecat.processors.frameworks.rtvi import RTVIObserver, RTVIProcessor
 from pipecat.runner.types import RunnerArguments
 from pipecat.runner.utils import create_transport
 from pipecat.services.cartesia.tts import CartesiaTTSService
@@ -87,8 +86,6 @@ async def run_bot(transport: BaseTransport, runner_args: RunnerArguments):
         ),
     )
 
-    rtvi = RTVIProcessor()
-
     pipeline = Pipeline(
         [
             transport.input(),  # Transport user input
@@ -108,13 +105,11 @@ async def run_bot(transport: BaseTransport, runner_args: RunnerArguments):
             enable_metrics=True,
             enable_usage_metrics=True,
         ),
-        observers=[RTVIObserver(rtvi)],
         idle_timeout_secs=runner_args.pipeline_idle_timeout_secs,
     )
 
-    @rtvi.event_handler("on_client_ready")
+    @task.rtvi.event_handler("on_client_ready")
     async def on_client_ready(rtvi):
-        await rtvi.set_bot_ready()
         # Kick off the conversation
         messages.append({"role": "system", "content": "Please introduce yourself to the user."})
         await task.queue_frames([LLMRunFrame()])
diff --git a/examples/foundational/46-video-processing.py b/examples/foundational/46-video-processing.py
index c35843eab..bb3d57f04 100644
--- a/examples/foundational/46-video-processing.py
+++ b/examples/foundational/46-video-processing.py
@@ -1,5 +1,5 @@
 #
-# Copyright (c) 2025, Daily
+# Copyright (c) 2024-2026, Daily
 #
 # SPDX-License-Identifier: BSD 2-Clause License
 #
@@ -22,7 +22,6 @@ from pipecat.processors.aggregators.llm_response_universal import (
     LLMUserAggregatorParams,
 )
 from pipecat.processors.frame_processor import FrameDirection, FrameProcessor
-from pipecat.processors.frameworks.rtvi import RTVIObserver, RTVIProcessor
 from pipecat.runner.types import RunnerArguments
 from pipecat.runner.utils import create_transport
 from pipecat.services.google.gemini_live.llm import GeminiLiveLLMService
@@ -125,14 +124,10 @@ async def run_bot(pipecat_transport):
         ),
     )
 
-    # RTVI events for Pipecat client UI
-    rtvi = RTVIProcessor()
-
     pipeline = Pipeline(
         [
             pipecat_transport.input(),
             user_aggregator,
-            rtvi,
             llm,  # LLM
             EdgeDetectionProcessor(
                 pipecat_transport._params.video_out_width,
@@ -149,13 +144,11 @@ async def run_bot(pipecat_transport):
             enable_metrics=True,
             enable_usage_metrics=True,
         ),
-        observers=[RTVIObserver(rtvi)],
     )
 
-    @rtvi.event_handler("on_client_ready")
+    @task.rtvi.event_handler("on_client_ready")
     async def on_client_ready(rtvi):
         logger.info("Pipecat client ready.")
-        await rtvi.set_bot_ready()
         # Kick off the conversation.
         await task.queue_frames([LLMRunFrame()])
 

From 124a3c35afa0c52e1dedb543790de7d6f34919e7 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Aleix=20Conchillo=20Flaqu=C3=A9?= <aleix@daily.co>
Date: Wed, 21 Jan 2026 14:25:25 -0800
Subject: [PATCH 32/51] RTVIObserver: don't handle some frames direction

---
 src/pipecat/processors/frameworks/rtvi.py | 2 --
 1 file changed, 2 deletions(-)

diff --git a/src/pipecat/processors/frameworks/rtvi.py b/src/pipecat/processors/frameworks/rtvi.py
index ee25af839..87ad556d9 100644
--- a/src/pipecat/processors/frameworks/rtvi.py
+++ b/src/pipecat/processors/frameworks/rtvi.py
@@ -1100,13 +1100,11 @@ class RTVIObserver(BaseObserver):
 
         if (
             isinstance(frame, (UserStartedSpeakingFrame, UserStoppedSpeakingFrame))
-            and (direction == FrameDirection.DOWNSTREAM)
             and self._params.user_speaking_enabled
         ):
             await self._handle_interruptions(frame)
         elif (
             isinstance(frame, (BotStartedSpeakingFrame, BotStoppedSpeakingFrame))
-            and (direction == FrameDirection.UPSTREAM)
             and self._params.bot_speaking_enabled
         ):
             await self._handle_bot_speaking(frame)

From 0ee11ad3335743dfe9b01c4a1798aee89874fd68 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Aleix=20Conchillo=20Flaqu=C3=A9?= <aleix@daily.co>
Date: Wed, 21 Jan 2026 14:31:04 -0800
Subject: [PATCH 33/51] tests: disable RTVI in tests by default

---
 src/pipecat/tests/utils.py | 11 +++++++----
 1 file changed, 7 insertions(+), 4 deletions(-)

diff --git a/src/pipecat/tests/utils.py b/src/pipecat/tests/utils.py
index ac5739829..356fb8545 100644
--- a/src/pipecat/tests/utils.py
+++ b/src/pipecat/tests/utils.py
@@ -123,9 +123,10 @@ class QueuedFrameProcessor(FrameProcessor):
 async def run_test(
     processor: FrameProcessor,
     *,
-    frames_to_send: Sequence[Frame],
+    enable_rtvi: bool = False,
     expected_down_frames: Optional[Sequence[type]] = None,
     expected_up_frames: Optional[Sequence[type]] = None,
+    frames_to_send: Sequence[Frame],
     ignore_start: bool = True,
     observers: Optional[List[BaseObserver]] = None,
     pipeline_params: Optional[PipelineParams] = None,
@@ -139,9 +140,10 @@ async def run_test(
 
     Args:
         processor: The frame processor to test.
-        frames_to_send: Sequence of frames to send through the processor.
+        enable_rtvi: Whether RTVI should be enabled in this test.
         expected_down_frames: Expected frame types flowing downstream (optional).
         expected_up_frames: Expected frame types flowing upstream (optional).
+        frames_to_send: Sequence of frames to send through the processor.
         ignore_start: Whether to ignore StartFrames in frame validation.
         observers: Optional list of observers to attach to the pipeline.
         pipeline_params: Optional pipeline parameters.
@@ -173,9 +175,10 @@ async def run_test(
 
     task = PipelineTask(
         pipeline,
-        params=pipeline_params,
-        observers=observers,
         cancel_on_idle_timeout=False,
+        enable_rtvi=enable_rtvi,
+        observers=observers,
+        params=pipeline_params,
     )
 
     async def push_frames():

From 9e8f8b45c682633ac354149f5c4ac6f93fc33925 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Aleix=20Conchillo=20Flaqu=C3=A9?= <aleix@daily.co>
Date: Wed, 21 Jan 2026 14:39:57 -0800
Subject: [PATCH 34/51] added changelog files for #3519

---
 changelog/3519.added.2.md | 1 +
 changelog/3519.added.3.md | 1 +
 changelog/3519.added.md   | 1 +
 changelog/3519.fixed.2.md | 1 +
 changelog/3519.fixed.md   | 1 +
 5 files changed, 5 insertions(+)
 create mode 100644 changelog/3519.added.2.md
 create mode 100644 changelog/3519.added.3.md
 create mode 100644 changelog/3519.added.md
 create mode 100644 changelog/3519.fixed.2.md
 create mode 100644 changelog/3519.fixed.md

diff --git a/changelog/3519.added.2.md b/changelog/3519.added.2.md
new file mode 100644
index 000000000..03e2372ad
--- /dev/null
+++ b/changelog/3519.added.2.md
@@ -0,0 +1 @@
+- Added `RTVIProcessor.create_rtvi_observer()` factory method for creating RTVI observers.
diff --git a/changelog/3519.added.3.md b/changelog/3519.added.3.md
new file mode 100644
index 000000000..7ea1e638c
--- /dev/null
+++ b/changelog/3519.added.3.md
@@ -0,0 +1 @@
+- Added `FrameProcessor.broadcast_frame_instance(frame)` method to broadcast a frame instance by extracting its fields and creating new instances for each direction.
diff --git a/changelog/3519.added.md b/changelog/3519.added.md
new file mode 100644
index 000000000..5ed2bd522
--- /dev/null
+++ b/changelog/3519.added.md
@@ -0,0 +1 @@
+- `PipelineTask` now automatically adds `RTVIProcessor` and registers `RTVIObserver` when `enable_rtvi=True` (default), simplifying pipeline setup.
diff --git a/changelog/3519.fixed.2.md b/changelog/3519.fixed.2.md
new file mode 100644
index 000000000..bcd384f7f
--- /dev/null
+++ b/changelog/3519.fixed.2.md
@@ -0,0 +1 @@
+- Fixed `FrameProcessor.broadcast_frame()` to deep copy kwargs, preventing shared mutable references between the downstream and upstream frame instances.
diff --git a/changelog/3519.fixed.md b/changelog/3519.fixed.md
new file mode 100644
index 000000000..cabaa5a3e
--- /dev/null
+++ b/changelog/3519.fixed.md
@@ -0,0 +1 @@
+- Transports now properly broadcast `InputTransportMessageFrame` frames both upstream and downstream instead of only pushing downstream.

From 7aa7b86aed78f0c5a1be44c90b211ec2c4b6a37d Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Aleix=20Conchillo=20Flaqu=C3=A9?= <aleix@daily.co>
Date: Wed, 21 Jan 2026 18:43:04 -0800
Subject: [PATCH 35/51] claude: initial /changelog skill

---
 .claude/skills/changelog/SKILL.md | 40 +++++++++++++++++++++++++++++++
 1 file changed, 40 insertions(+)
 create mode 100644 .claude/skills/changelog/SKILL.md

diff --git a/.claude/skills/changelog/SKILL.md b/.claude/skills/changelog/SKILL.md
new file mode 100644
index 000000000..0feb6b92c
--- /dev/null
+++ b/.claude/skills/changelog/SKILL.md
@@ -0,0 +1,40 @@
+---
+name: changelog
+description: Create changelog files for important commits in a PR
+---
+
+Create changelog files for the important commits in this PR. The PR number is provided as an argument.
+
+## Instructions
+
+1. First, check what commits are on the current branch compared to main:
+   ```
+   git log main..HEAD --oneline
+   ```
+
+2. For each significant change, create a changelog file in the `changelog/` folder using the format:
+   - `{PR_NUMBER}.added.md` - for new features
+   - `{PR_NUMBER}.added.2.md`, `{PR_NUMBER}.added.3.md` - for additional new features
+   - `{PR_NUMBER}.changed.md` - for changes to existing functionality
+   - `{PR_NUMBER}.fixed.md` - for bug fixes
+   - `{PR_NUMBER}.deprecated.md` - for deprecations
+
+3. Each changelog file should at least contain a main single line starting with `- ` followed by a clear description of the change.
+
+4. If the change is complicated, changelog files can have indented lines after the main line with additional details or code samples.
+
+5. Use ⚠️ emoji prefix for breaking changes.
+
+## Example
+
+For PR #3519 with a new feature and a bug fix:
+
+`changelog/3519.added.md`:
+```
+- Added `SomeNewFeature` for doing something useful.
+```
+
+`changelog/3519.fixed.md`:
+```
+- Fixed an issue where something was not working correctly.
+```

From 6e44a2ab493371e6c28ca823eb118abf3696bcac Mon Sep 17 00:00:00 2001
From: Om Chauhan <omchauhan64408@gmail.com>
Date: Wed, 21 Jan 2026 08:14:48 +0530
Subject: [PATCH 36/51] feat(task): add additive filter methods for frame
 monitoring

---
 src/pipecat/pipeline/task.py | 42 ++++++++++++++++++++++++++++++++++++
 1 file changed, 42 insertions(+)

diff --git a/src/pipecat/pipeline/task.py b/src/pipecat/pipeline/task.py
index 9a2f1d33d..20b4adf36 100644
--- a/src/pipecat/pipeline/task.py
+++ b/src/pipecat/pipeline/task.py
@@ -427,6 +427,24 @@ class PipelineTask(BasePipelineTask):
             raise Exception(f"{self} RTVI is not enabled.")
         return self._rtvi
 
+    @property
+    def reached_upstream_types(self) -> Tuple[Type[Frame], ...]:
+        """Get the currently configured upstream frame type filters.
+
+        Returns:
+            Tuple of frame types that trigger the on_frame_reached_upstream event.
+        """
+        return self._reached_upstream_types
+
+    @property
+    def reached_downstream_types(self) -> Tuple[Type[Frame], ...]:
+        """Get the currently configured downstream frame type filters.
+
+        Returns:
+            Tuple of frame types that trigger the on_frame_reached_downstream event.
+        """
+        return self._reached_downstream_types
+
     def event_handler(self, event_name: str):
         """Decorator for registering event handlers.
 
@@ -480,6 +498,30 @@ class PipelineTask(BasePipelineTask):
         """
         self._reached_downstream_types = types
 
+    def add_reached_upstream_filter(self, types: Tuple[Type[Frame], ...]):
+        """Add frame types to trigger the on_frame_reached_upstream event.
+
+        Args:
+            types: Tuple of frame types to add to upstream monitoring.
+        """
+        if not types:
+            return
+        current = set(self._reached_upstream_types)
+        current.update(types)
+        self._reached_upstream_types = tuple(current)
+
+    def add_reached_downstream_filter(self, types: Tuple[Type[Frame], ...]):
+        """Add frame types to trigger the on_frame_reached_downstream event.
+
+        Args:
+            types: Tuple of frame types to add to downstream monitoring.
+        """
+        if not types:
+            return
+        current = set(self._reached_downstream_types)
+        current.update(types)
+        self._reached_downstream_types = tuple(current)
+
     def has_finished(self) -> bool:
         """Check if the pipeline task has finished execution.
 

From 9c0bf892472cf0d8b92f87d7547342156856fca2 Mon Sep 17 00:00:00 2001
From: Om Chauhan <omchauhan64408@gmail.com>
Date: Wed, 21 Jan 2026 08:36:20 +0530
Subject: [PATCH 37/51] added changelog

---
 changelog/3510.added.2.md | 1 +
 changelog/3510.added.md   | 1 +
 2 files changed, 2 insertions(+)
 create mode 100644 changelog/3510.added.2.md
 create mode 100644 changelog/3510.added.md

diff --git a/changelog/3510.added.2.md b/changelog/3510.added.2.md
new file mode 100644
index 000000000..e03b81ee3
--- /dev/null
+++ b/changelog/3510.added.2.md
@@ -0,0 +1 @@
+- Added `add_reached_upstream_filter()` and `add_reached_downstream_filter()` methods to `PipelineTask` for appending frame types.
\ No newline at end of file
diff --git a/changelog/3510.added.md b/changelog/3510.added.md
new file mode 100644
index 000000000..93d9ef179
--- /dev/null
+++ b/changelog/3510.added.md
@@ -0,0 +1 @@
+- Added `reached_upstream_types` and `reached_downstream_types` read-only properties to `PipelineTask` for inspecting current frame filters.
\ No newline at end of file

From 87c12f3098c4176788a8090c1890bdf7dbd00bf0 Mon Sep 17 00:00:00 2001
From: Om Chauhan <omchauhan64408@gmail.com>
Date: Thu, 22 Jan 2026 08:39:31 +0530
Subject: [PATCH 38/51] changed frame filter storage type from tuples to sets

---
 changelog/3510.changed.3.md  |  1 +
 src/pipecat/pipeline/task.py | 30 +++++++++++-------------------
 2 files changed, 12 insertions(+), 19 deletions(-)
 create mode 100644 changelog/3510.changed.3.md

diff --git a/changelog/3510.changed.3.md b/changelog/3510.changed.3.md
new file mode 100644
index 000000000..ac299f143
--- /dev/null
+++ b/changelog/3510.changed.3.md
@@ -0,0 +1 @@
+- Changed frame filter storage from tuples to sets in `PipelineTask`.
diff --git a/src/pipecat/pipeline/task.py b/src/pipecat/pipeline/task.py
index 20b4adf36..60df77ff9 100644
--- a/src/pipecat/pipeline/task.py
+++ b/src/pipecat/pipeline/task.py
@@ -15,7 +15,7 @@ import asyncio
 import importlib.util
 import os
 from pathlib import Path
-from typing import Any, AsyncIterable, Dict, Iterable, List, Optional, Tuple, Type
+from typing import Any, AsyncIterable, Dict, Iterable, List, Optional, Set, Tuple, Type
 
 from loguru import logger
 from pydantic import BaseModel, ConfigDict, Field
@@ -366,8 +366,8 @@ class PipelineTask(BasePipelineTask):
         # in. This is mainly for efficiency reason because each event handler
         # creates a task and most likely you only care about one or two frame
         # types.
-        self._reached_upstream_types: Tuple[Type[Frame], ...] = ()
-        self._reached_downstream_types: Tuple[Type[Frame], ...] = ()
+        self._reached_upstream_types: Set[Type[Frame]] = set()
+        self._reached_downstream_types: Set[Type[Frame]] = set()
         self._register_event_handler("on_frame_reached_upstream")
         self._register_event_handler("on_frame_reached_downstream")
         self._register_event_handler("on_idle_timeout")
@@ -434,7 +434,7 @@ class PipelineTask(BasePipelineTask):
         Returns:
             Tuple of frame types that trigger the on_frame_reached_upstream event.
         """
-        return self._reached_upstream_types
+        return tuple(self._reached_upstream_types)
 
     @property
     def reached_downstream_types(self) -> Tuple[Type[Frame], ...]:
@@ -443,7 +443,7 @@ class PipelineTask(BasePipelineTask):
         Returns:
             Tuple of frame types that trigger the on_frame_reached_downstream event.
         """
-        return self._reached_downstream_types
+        return tuple(self._reached_downstream_types)
 
     def event_handler(self, event_name: str):
         """Decorator for registering event handlers.
@@ -488,7 +488,7 @@ class PipelineTask(BasePipelineTask):
         Args:
             types: Tuple of frame types to monitor for upstream events.
         """
-        self._reached_upstream_types = types
+        self._reached_upstream_types = set(types)
 
     def set_reached_downstream_filter(self, types: Tuple[Type[Frame], ...]):
         """Set which frame types trigger the on_frame_reached_downstream event.
@@ -496,7 +496,7 @@ class PipelineTask(BasePipelineTask):
         Args:
             types: Tuple of frame types to monitor for downstream events.
         """
-        self._reached_downstream_types = types
+        self._reached_downstream_types = set(types)
 
     def add_reached_upstream_filter(self, types: Tuple[Type[Frame], ...]):
         """Add frame types to trigger the on_frame_reached_upstream event.
@@ -504,11 +504,7 @@ class PipelineTask(BasePipelineTask):
         Args:
             types: Tuple of frame types to add to upstream monitoring.
         """
-        if not types:
-            return
-        current = set(self._reached_upstream_types)
-        current.update(types)
-        self._reached_upstream_types = tuple(current)
+        self._reached_upstream_types.update(types)
 
     def add_reached_downstream_filter(self, types: Tuple[Type[Frame], ...]):
         """Add frame types to trigger the on_frame_reached_downstream event.
@@ -516,11 +512,7 @@ class PipelineTask(BasePipelineTask):
         Args:
             types: Tuple of frame types to add to downstream monitoring.
         """
-        if not types:
-            return
-        current = set(self._reached_downstream_types)
-        current.update(types)
-        self._reached_downstream_types = tuple(current)
+        self._reached_downstream_types.update(types)
 
     def has_finished(self) -> bool:
         """Check if the pipeline task has finished execution.
@@ -820,7 +812,7 @@ class PipelineTask(BasePipelineTask):
         pipeline to be stopped (e.g. EndTaskFrame) in which case we would send
         an EndFrame down the pipeline.
         """
-        if isinstance(frame, self._reached_upstream_types):
+        if isinstance(frame, tuple(self._reached_upstream_types)):
             await self._call_event_handler("on_frame_reached_upstream", frame)
 
         if isinstance(frame, EndTaskFrame):
@@ -859,7 +851,7 @@ class PipelineTask(BasePipelineTask):
         processors have handled the EndFrame and therefore we can exit the task
         cleanly.
         """
-        if isinstance(frame, self._reached_downstream_types):
+        if isinstance(frame, tuple(self._reached_downstream_types)):
             await self._call_event_handler("on_frame_reached_downstream", frame)
 
         if isinstance(frame, StartFrame):

From 8f05d95f50638a400b10a0fdb1df3ad10002f76c Mon Sep 17 00:00:00 2001
From: James Hush <james@daily.co>
Date: Thu, 22 Jan 2026 11:31:07 +0800
Subject: [PATCH 39/51] feat: add video_out_codec parameter for DailyTransport
 (#3520)

* feat: add video_out_codec parameter for DailyTransport

Add video_out_codec parameter to TransportParams allowing configuration
of the preferred video codec (VP8, H264, H265) for video output.

When set, this passes the preferredCodec option to Daily's
VideoPublishingSettings during the join operation.

* chore: move video_out_codec parameter to changelog folder (#3522)

* Initial plan

* Move video_out_codec parameter to changelog/3520.added.md

Co-authored-by: jamsea <614910+jamsea@users.noreply.github.com>

* Revert all CHANGELOG.md changes, keep only changelog/3520.added.md

Co-authored-by: jamsea <614910+jamsea@users.noreply.github.com>

---------

Co-authored-by: copilot-swe-agent[bot] <198982749+Copilot@users.noreply.github.com>
Co-authored-by: jamsea <614910+jamsea@users.noreply.github.com>

---------

Co-authored-by: Copilot <198982749+Copilot@users.noreply.github.com>
Co-authored-by: jamsea <614910+jamsea@users.noreply.github.com>
---
 application-logs.log                      | 610 ++++++++++++++++++++++
 changelog/3520.added.md                   |   1 +
 src/pipecat/transports/base_transport.py  |   2 +
 src/pipecat/transports/daily/transport.py |   5 +
 4 files changed, 618 insertions(+)
 create mode 100644 application-logs.log
 create mode 100644 changelog/3520.added.md

diff --git a/application-logs.log b/application-logs.log
new file mode 100644
index 000000000..32c97c288
--- /dev/null
+++ b/application-logs.log
@@ -0,0 +1,610 @@
+[ai] [2026-01-16 11:04:04 -0800] [59087] [INFO] 127.0.0.1:53364 OPTIONS /ai2/start-voice-assistant 1.1 200 0 9044
+[ai] INFO:sunsama_ai.server:Starting a call
+[ai] INFO:sunsama_ai.server:CallId: None CallDomain: None DialoutNumber: None DialinNumber: None DialedNumber: None Flow: braindump
+[ai] INFO:sunsama_ai.server:
+[ai]
+[ai]
+[ai]
+[ai] INFO:sunsama_ai.server:Call recording initialized for voice_session_id: 696a8ba4de74410788876fbb
+[ai] INFO:sunsama_ai.voice_assistant.bot_runner:Creating new room...
+[ai] INFO:sunsama_ai.voice_assistant.bot_runner:Daily room: https://cloud-99400c183d52498c940fe96f45c05b44.daily.co/teWpRxCQnkMHYi7Q3lqh
+[ai] INFO:sunsama_ai.voice_assistant.bot_runner:Starting bot call...
+[ai] INFO:sunsama_ai.voice_assistant.bot_runner:Bot call started
+[ai] [2026-01-16 11:04:04 -0800] [59087] [INFO] 127.0.0.1:53364 POST /ai2/start-voice-assistant 1.1 200 356 358242
+[voice] INFO:sunsama_voice.bot:Bot process initialized None https://cloud-99400c183d52498c940fe96f45c05b44.daily.co/teWpRxCQnkMHYi7Q3lqh eyJhbGciOiJIUzI1NiIsInR5cCI6IkpXVCJ9.eyJyIjoidGVXcFJ4Q1Fua01IWWk3UTNscWgiLCJlanQiOmZhbHNlLCJleHAiOjE3Njg1OTA1NDQsIm8iOnRydWUsImQiOiI5OWJiNzdhNS0wYjZkLTRkOWMtOTkxMC1jMDcwNmEwZTA0ZDMiLCJpYXQiOjE3Njg1OTAyNDR9.npzlQ2ncZ7zemVHacAgerEYBSchiEM-VYxKf3t8Z5A0
+[voice] INFO:sunsama_voice.bot:{'roomUrl': 'https://cloud-99400c183d52498c940fe96f45c05b44.daily.co/teWpRxCQnkMHYi7Q3lqh', 'token': 'eyJhbGciOiJIUzI1NiIsInR5cCI6IkpXVCJ9.eyJyIjoidGVXcFJ4Q1Fua01IWWk3UTNscWgiLCJlanQiOmZhbHNlLCJleHAiOjE3Njg1OTA1NDQsIm8iOnRydWUsImQiOiI5OWJiNzdhNS0wYjZkLTRkOWMtOTkxMC1jMDcwNmEwZTA0ZDMiLCJpYXQiOjE3Njg1OTAyNDR9.npzlQ2ncZ7zemVHacAgerEYBSchiEM-VYxKf3t8Z5A0', 'callId': None, 'callDomain': None, 'detectVoicemail': False, 'dialoutNumber': None, 'dialinNumber': None, 'dialedNumber': None, 'authToken': 'eyJhbGciOiJIUzI1NiIsInR5cCI6IkpXVCJ9.eyJ1c2VySWQiOiI2NzFhYzVlNDAyYTMwMmNmNmVmMzA4OGUiLCJzZXNzaW9uSWQiOiIxOTVmNDI0NC05YjFmLTRlYmItOTMwZC0wZTExNTAwNzZkYzkiLCJzZXNzaW9uTm9uY2UiOiJhY2ZlZTc2OS1jZGU2LTRkMWYtYTU5Ni0xMzk0MTRkMjUxNjAiLCJncm91cElkIjoiNjcxYWM1ZTkwMmEzMDJjZjZlZjMwODkwIiwiZ3JvdXBVcmxQYXRoIjoiMTcyOTgwNzg0OTAzNjUxOTIiLCJpYXQiOjE3Njg1OTAyMTYsImV4cCI6MTc3MTE4MjIxNn0.HHSuewYqBRl-IZkdiWcEJyiD6akgbtwnOIQ4Js71bZI', 'flow': 'braindump', 'flowOptions': {}, 'voiceSessionId': '696a8ba4de74410788876fbb'}
+[voice] /Users/zanemccaig/workspaces/sunsama/sunsama/sunsama-voice/venv/lib/python3.13/site-packages/pipecat/services/google/llm.py:281: DeprecationWarning: OpenAILLMContext is deprecated and will be removed in a future version. Use the universal LLMContext and LLMContextAggregatorPair instead. See OpenAILLMContext docstring for migration guide.
+[voice]   super().__init__(messages=messages, tools=tools, tool_choice=tool_choice)
+[voice] INFO:sunsama_voice.mcp_websocket_client:Sending message: {'method': 'initialize', 'params': {'protocolVersion': '2025-06-18', 'capabilities': {}, 'clientInfo': {'name': 'mcp', 'version': '0.1.0'}}, 'jsonrpc': '2.0', 'id': 0}
+[api-local] [api] 2026-01-16T19:04:04.835Z info: Setting up user subscription for MCP context: 671ac5e402a302cf6ef3088e
+[api-local] [api] Registered handler for channel "user-updates-671ac5e402a302cf6ef3088e" for event "userUpdated"
+[api-local] [api] 2026-01-16T19:04:04.835Z info: Successfully subscribed to user updates for MCP context: 671ac5e402a302cf6ef3088e
+[api-local] [api] Starting WebSocket connection...
+[voice] INFO:sunsama_voice.mcp_websocket_client:Received message: {"result":{"protocolVersion":"2025-06-18","capabilities":{"tools":{"listChanged":true},"prompts":{"listChanged":true},"completions":{},"resources":{"listChanged":true}},"serverInfo":{"name":"Sunsama MCP Server","version":"0.1.0"}},"jsonrpc":"2.0","id":0}
+[voice] INFO:sunsama_voice.mcp_websocket_client:Sending message: {'method': 'notifications/initialized', 'jsonrpc': '2.0'}
+[voice] INFO:sunsama_voice.bot:Timer: MCP Session created in 0.0885460376739502 seconds
+[voice] DEBUG:sunsama_voice.bot:Starting bot with room_url: https://cloud-99400c183d52498c940fe96f45c05b44.daily.co/teWpRxCQnkMHYi7Q3lqh
+[voice] /Users/zanemccaig/workspaces/sunsama/sunsama/sunsama-voice/sunsama_voice/models.py:254: DeprecationWarning: FalSmartTurnAnalyzer is deprecated and will be removed in a future version. Use LocalSmartTurnAnalyzerV3 instead.
+[voice]   turn_analyzer = FalSmartTurnAnalyzer(
+[voice] DEBUG:pipecat.audio.vad.silero:Loading Silero VAD model...
+[voice] DEBUG:pipecat.audio.vad.silero:Loaded Silero VAD
+[voice] /Users/zanemccaig/workspaces/sunsama/sunsama/sunsama-voice/venv/lib/python3.13/site-packages/pipecat/processors/aggregators/llm_response.py:326: DeprecationWarning: GoogleUserContextAggregator (likely created with create_context_aggregator()) is deprecated and will be removed in a future version. Use the universal LLMContext and LLMContextAggregatorPair instead. See OpenAILLMContext docstring for migration guide.
+[voice]   super().__init__(**kwargs)
+[voice] /Users/zanemccaig/workspaces/sunsama/sunsama/sunsama-voice/venv/lib/python3.13/site-packages/pipecat/processors/aggregators/llm_response.py:326: DeprecationWarning: GoogleAssistantContextAggregator (likely created with create_context_aggregator()) is deprecated and will be removed in a future version. Use the universal LLMContext and LLMContextAggregatorPair instead. See OpenAILLMContext docstring for migration guide.
+[voice]   super().__init__(**kwargs)
+[voice] /Users/zanemccaig/workspaces/sunsama/sunsama/sunsama-voice/venv/lib/python3.13/site-packages/pipecat/transports/base_input.py:135: DeprecationWarning: Parameter 'turn_analyzer' is deprecated, use `LLMUserAggregator`'s new `user_turn_strategies` parameter instead.
+[voice]   warnings.warn(
+[voice] DEBUG:pipecat.processors.frame_processor:Linking Pipeline#0::Source -> DailyInputTransport#0
+[voice] DEBUG:pipecat.processors.frame_processor:Linking DailyInputTransport#0 -> RTVIProcessor#0
+[voice] DEBUG:pipecat.processors.frame_processor:Linking RTVIProcessor#0 -> ClientMessageProcessor#0
+[voice] DEBUG:pipecat.processors.frame_processor:Linking ClientMessageProcessor#0 -> CustomUserIdleProcessor#0
+[voice] DEBUG:pipecat.processors.frame_processor:Linking CustomUserIdleProcessor#0 -> GoogleUserContextAggregator#0
+[voice] DEBUG:pipecat.processors.frame_processor:Linking GoogleUserContextAggregator#0 -> GoogleLLMService#0
+[voice] DEBUG:pipecat.processors.frame_processor:Linking GoogleLLMService#0 -> CartesiaTTSService#1
+[voice] DEBUG:pipecat.processors.frame_processor:Linking CartesiaTTSService#1 -> DailyOutputTransport#0
+[voice] DEBUG:pipecat.processors.frame_processor:Linking DailyOutputTransport#0 -> GoogleAssistantContextAggregator#0
+[voice] DEBUG:pipecat.processors.frame_processor:Linking GoogleAssistantContextAggregator#0 -> Pipeline#0::Sink
+[voice] /Users/zanemccaig/workspaces/sunsama/sunsama/sunsama-voice/venv/lib/python3.13/site-packages/pipecat/pipeline/task.py:273: DeprecationWarning: Field 'observers' is deprecated, use the 'observers' parameter instead.
+[voice]   warnings.warn(
+[voice] DEBUG:pipecat.processors.frame_processor:Linking PipelineTask#0::Source -> Pipeline#0
+[voice] DEBUG:pipecat.processors.frame_processor:Linking Pipeline#0 -> PipelineTask#0::Sink
+[voice] INFO:sunsama_voice.bot:Timer: Pipeline created in 0.1510629653930664 seconds
+[voice] INFO:sunsama_voice.mcp:Fetching resource sunsama://user/authentication with args: None	sunsama://user/authentication
+[voice] INFO:sunsama_voice.mcp_websocket_client:Sending message: {'method': 'resources/read', 'params': {'uri': 'sunsama://user/authentication'}, 'jsonrpc': '2.0', 'id': 1}
+[voice] INFO:sunsama_voice.mcp_websocket_client:Received message: {"result":{"contents":[{"uri":"sunsama://user/authentication","mimeType":"text/plain","text":"authenticated"}]},"jsonrpc":"2.0","id":1}
+[voice] INFO:sunsama_voice.mcp:Resource sunsama://user/authentication result: meta=None contents=[TextResourceContents(uri=AnyUrl('sunsama://user/authentication'), mimeType='text/plain', meta=None, text='authenticated')]
+[voice] INFO:sunsama_voice.mcp:Resource sunsama://user/authentication result: uri=AnyUrl('sunsama://user/authentication') mimeType='text/plain' meta=None text='authenticated'
+[voice] INFO:sunsama_voice.mcp:Handling resource update: uri sunsama://user/authentication authenticated 2026-01-16 11:04:04.903462 error=None status=None
+[voice] INFO:sunsama_voice.mcp_websocket_client:Sending message: {'method': 'resources/list', 'jsonrpc': '2.0', 'id': 2}
+[voice] INFO:sunsama_voice.mcp_websocket_client:Received message: {"result":{"resources":[{"uri":"sunsama://user/authentication","name":"get_user_authentication_status","description":"Retrieves the user's authentication status.","mimeType":"application/json"},{"uri":"sunsama://me","name":"get_user_info","description":"Critical information about the user like their name, timezone, and the current time where they're calling in from. Use this to personalize the conversation.","mimeType":"application/json"},{"uri":"sunsama://email/accounts","name":"list_email_accounts","description":"Lists email accounts the user has connected to Sunsama.","mimeType":"application/json"},{"uri":"sunsama://daily-highlight","name":"get_last_daily_highlight","description":"Retrieves their journal entry from their last workday in Markdown format.","mimeType":"text/markdown"},{"uri":"sunsama://backlog/folders","name":"get_backlog_folders","description":"Retrieves the user's backlog folders.","mimeType":"application/json"},{"uri":"sunsama://active-timer","name":"get_active_timer","description":"Retrieves information about the currently active timer, including which task has the timer running and when it was started.","mimeType":"application/json"},{"uri":"sunsama://ui/available-actions","name":"get_available_ui_actions","description":"Retrieves the list of available user interface actions that can be triggered by the assistant. This list may change based on the user's current context and permissions. Each action includes its actionId, type, and context information.","mimeType":"application/json"}]},"jsonrpc":"2.0","id":2}
+[voice] INFO:sunsama_voice.mcp:Resource updated: uri sunsama://user/authentication
+[voice] INFO:sunsama_voice.bot:Timer: Authentication status checked in 0.15551996231079102 seconds
+[voice] INFO:sunsama_voice.mcp_websocket_client:Sending message: {'method': 'prompts/get', 'params': {'name': 'braindump', 'arguments': {}}, 'jsonrpc': '2.0', 'id': 3}
+[api-local] [api] preFetchResources:get_backlog_folders:317522: 1.969ms
+[api-local] [api] preFetchResources:get_user_info:436506: 6.557ms
+[api-local] [api] preFetchResources:BRAINDUMP: 6.712ms
+[voice] INFO:sunsama_voice.mcp_websocket_client:Received message: {"result":{"messages":[{"role":"user","content":{"type":"text","text":"You are Sunny, the personal assistant for a busy professional. You are part of Sunsama.\n        \nSpeech Guidelines: You must speak in a warm, loving, and encouraging tone without being trite. Speak quickly and add emotion. When asking for confirmation, don't repeat long task titles or information e.g. \"Got it\" instead \"Got it, I'll add a task for two hours for you to work on the presentation today\"\n\nYour personality and philosophy:\n- You don't buy into hustle culture. For you, less is more. It's better to do fewer things well and to do them with deep focus. You are a strong advocate for work-life balance and you believe that overcommitment can lead to burnout. You always remind the user when they are overcommitting themselves.\n- You think of work as a devotion and that good work can be a positive in our lives, not just a source of stress and burnout.\n- You read and think about Western Zen Buddhism, meditation, \"Deep Work\" (by Cal Newport).\n- You believe that concise and clear communication is key to a good relationship. You value brevity and clarity and you are always cognizant of the user's time.\n- You respond to the user using quick, short sentences. Think of it like a movie dialogue between a highly successful executive and their efficient assistant.\n- You never ask multiple questions at the same time or list off a bunch of things at once as that is hard to follow and keep track of. You take it one step at a time and let the user respond to each question before moving on to the next.\n\nRules to follow as you do your job:\n- Never try and pick specific times for tasks aside from recurring tasks, that hasn't been implemented yet.\n- Do not regurgitate what the user just said, that's very annoying. Just say \"Got it\" or \"Alright\", etc.\n- Always use units that are more intuitive. For example:\n    - Don't say \"270 minutes\", say \"4 and a half hours\".\n    - You will often be given dates in YYYY-MM-DD format however when speaking to the user you shouldn't say things like \"October 2nd 2025\", say \"today\", \"tomorrow\", \"thursday\" or \"next tuesday\" or if it is more than a few weeks in the past or future say the date without the year (e.g. \"December 14th\"). Only mention the year if it is not this year and not within the next month.\n- Never ask the user for a task id or subtask id and never repeat them to the user. Those are internal and only there to help you do your job.\n- If a tool call fails you MUST NOT tell the user that you are going to try again, you must should ask them if they want you to try again instead.\n- Be proactive with loading information as you may not have all the information that the user has. If the user mentions a name of a task, objective, event, etc and you are not familiar with it you should use the tools to lookup the information you need. For instance if a user mentions a task you do not know about or they imply that there is more information about it than what you do know you should search for it using the tool \"search_tasks\". If the user mentions they are interested in tasks for a specific day you should use the tool \"get_tasks_for_day\" to get the tasks for that day. Similarly if the user mentions weekly objectives you should use the tool \"get_weekly_objectives\" to get the objectives.\n- When listing tasks try to be concise. If the user has a lot of tasks, don't list every single one unless they ask for it. You should instead try to summarize the tasks in a concise manner.\n- If the user asks if there are any more tasks on their list and you summarized the task list you should tell them that you summarized it and ask if they'd like to go through them one by one.\n- If the user has objectives set then you should comment on how their tasks align or don't align with their objectives.\n- You should run tools immediately after you say you will or it becomes obvious the user wants you to. Do not wait until later.\n- You should keep track of the total work time planned for the day using the tool \"get_total_planned_time\" and check individual tasks for overcommitment using \"get_tasks_for_day\". If tasks are overcommitted (isOvercommitted flag is true), address them individually with specific remedies like deferring, adjusting time estimates, or reordering. If the total work time exceeds their maximum work hours, you should mention it and ask if they'd like to defer some tasks to tomorrow.\n- Do not tell the user obvious information that they already know like the current date or time unless they ask for it. That kind of information is only provided to you to help you do your job.\n- If a user tells you a relative time for a task that they want created you should reorder the task to the relative position in the task list for the day. Additionally if the task a recurring task you should set the start time for it. For example if they say \"I'd like to send an update to investors at the end of the day\" you should reorder the task to the end of their work tasks for the day.\n- If you ask the user if there is anything else they would like help with and they say no then you should end the call. Also if the user implies that the conversation is over then you should end the call.\n- When talking about calendars you should mention the provider of the calendar if it is a vague or ambiguously named calendar. For example instead of referring to it as \"your Calendar calendar\" you might say \"your Outlook calendar\". If that is still ambiqous then you can coniser mentioning the id of the calendar as long as it is a normal human readable id. Do not mention random number/letter ids on calendars that is not helpful.\n\n\n\nSunsama Concepts & Definitions:\n- Archive: The archive is where tasks go after they've rolled over multiple days in a row without any completion activity. This is a distinct concept from the backlog or deleted tasks.\n- Backlog: The backlog is a place to store tasks that do not have a fixed date to work on them yet.\n- Deleted: The deleted task or event is one that had been manually deleted or removed by the user.\n- Task: A task is an item that a user wants to complete. It is either scheduled for a specific date or is in the backlog.\n- Event: An event is an item that a user has scheduled on their calendar. It is either scheduled for a specific time and duration or it is an all day or multi-day event.\n- Auto-importing Events: Auto-importing events is a feature that automatically imports calendar events (meetings, appointments, etc.) from the user's connected calendars into their daily task list. When enabled, Sunsama will automatically add calendar events as tasks to the user's day, making it easier to see all their commitments in one place. Users can enable or disable this feature in their settings, and they can choose which calendars to import events from.\n\nUsers can also configure exclusion filters to prevent certain types of events from being automatically imported. Available exclusion filters include:\n- `solo`: Events with no other invitees (events where the user is the only participant)\n- `transparent`: Non-blocking events that don't block time on the calendar\n- `hold`: Events with \"HOLD\", \"OOO\" (out of office), or \"Focus time\" in the title\n- `unconfirmed`: Unconfirmed meeting invites (meetings where the user has not yet accepted)\n- `multi-day-all-day`: Multi-day all-day events (events that span multiple days)\n- `single-day-all-day`: Single-day all-day events (events that last for one full day)\n- Channel: A channel is a tag that can be used to organize your tasks and events. Each task/event can be assigned to a single channel. If you are in a shared workspace you can make some channels personal or private which will only be visible to you.\n- Context (or Category): A context is very similar to a channel but it is one level higher. You can use a context to group similar or related channels together. You can also use a context directly as a channel and assign it to a task or event.\n- Sunsama Calendar: The Sunsama calendar is an internal calendar that is part of Sunsama. Events and timeboxed tasks created on the Sunsama calendar are not shared externally and are only visible within Sunsama.\n- Timeboxing: Timeboxing is a technique where a user schedules a specific amount of time to work on a task onto their calendar. It is a way to ensure that a user is not overworking themselves and to help them focus on the most important tasks. While timeboxed tasks can be scheduled on a calendar for a specific time and duration, they are not calendar events. Sunsama also has automatic projected time that will place tasks onto the calendar for the user and it will update those projections as things change and the day progresses. Users can pin tasks to the calendar to override the projections.\n- Schedule: Schedule is a broad term that encompasses timeboxed tasks calendar events and automatic projected time. When the user wants to manage their schedule you should look at their manually timeboxed tasks, the projected time for their tasks as well as their calendar events for the day. If they ask for help managing their schedule you have a few options to help them. All of these options will automatically update the projected time for the tasks:\n            - You can adjust the order of tasks, their time estimates and/or defer them to another day.\n            - You can move calendar events to a new time on the calendar.\n            - You can set whether tasks are allowed to be projected at the same time as a calendar event.\n            - You can timebox tasks to the calendar manually.\n- Objective: An objective is a goal that a user can set for the week. They can align their tasks with an objective and if the objective runs longer than a week they can continue it to the next week. Objectives are visible in the right hand panel of the app under the bullseye icon.\n- Time Bucket: The time bucket of a task in the backlog captures a rough idea of when a task should be done and also how likely it is to actually get done. Options: \"in the next two weeks\", \"in the next month\", \"in the next quarter\", \"in the next year\", \"someday\", \"never\". \n\n\n\nSunsama FAQ:\n- How can I use Sunny in the app?: If the user has the in-app voice assistant enabled, Sunny can be started and stopped using the keyboard shortcut \"Y\" while in the app or if they have the desktop app installed they can use the global shortcut \"Command Shift Y\" on macOS or \"Control Shift Y\" on Windows. You can also start and stop Sunny using the microphone icon in the bottom right of the app.\n- How can I use Sunny over the phone?: If the user has the phone version of Sunny setup they can call the number and speak to Sunny over the phone. You can find the phone number in the Sunsama settings. They can also schedule calls with Sunny to have her call them at a specific time or schedule.\n- What can you do?: Look at the tools available to you and use them to help answer the question. Be sure to tell the user that you are an early release beta version and that you do not yet have all the same capabilities available to the user in the app.\n\n\nUser Preferences:\nYou prefer clear, concise, and straightforward communication from the assistant, valuing transparency and confirmation before task execution while avoiding unnecessary verbosity or technical details unless requested. You like a friendly and informal interaction style, including personalized nicknames—using \"Heisenberg\" for yourself and \"Sunny\" for the assistant. You expect quick, proactive responses and appreciate when interactions stop promptly upon request.\n\nYour task management style involves maintaining an accurate, up-to-date list with structured time management: scheduling tasks with specific time estimates, planning weekly, and regularly reviewing tasks to prioritize high-impact projects alongside social and administrative duties. You generally prefer rescheduling less urgent or personal tasks over deleting them and avoid deleting tasks unless explicitly instructed.\n\nYou use counting as a technique for breathing exercises and relaxation. You also keep certain important tasks (e.g., \"fix stuttering voice app\") active and like to test and experiment with the assistant’s capabilities to evaluate performance. Your primary calendar for events is Google Calendar.\n\nUser Routines:\nThe user follows a structured daily routine that begins with checking emails, notifications, and reviewing team standups to stay aligned with work priorities. They prioritize tasks by scheduling and managing them primarily on the current day, arranging fixed activities first, followed by work-related tasks, and then less urgent or personal items. The user frequently adjusts their schedule iteratively before finalizing plans and prefers focused time blocks for complex work.\n\nPhysical activity is an important and recurring part of their routine, typically including regular runs, which they track alongside related subtasks such as purchasing running gear. Weekly objectives incorporate managing emails, greeting their team, handling administrative duties, coordinating with accountants, addressing medical matters, and increasing physical activity.\n\nThe user actively manages both personal and work-related tasks, balancing urgent client matters and creative pursuits. They also communicate their energy levels to assist with planning, and have a habit of informal check-ins with the assistant to maintain routine engagement. Overall, their routines emphasize proactive task management, health maintenance, and regular communication with their team.\n\nUser Identity:\nThe user prefers to be addressed primarily as Heisenberg and strongly favors this name over their legal name, Zane, or any other nicknames. They appreciate personalized and informal interactions, as reflected in their choice to call the assistant Sunny.\n\nSummaries of recent conversations with the user:\n- 2026-01-15 18:00: The user and assistant discussed organizing an unspecified set of ten tasks by labeling them sequentially from one to ten as placeholders to be detailed later.\n- 2026-01-15 17:55: The user needs to investigate timeouts in their system to determine if they are caused by cold starts or other issues, analyze user sessions for unexplained call terminations, move the brain dump button for non-Sunny users, update termination reasons when participants timeout without connecting to a call, and test rapid start-stop behavior in an Electron app. These tasks were added to their backlog for the next two weeks.\n- 2026-01-14 21:12: The user and the assistant exchanged greetings, with the assistant addressing the user as 'Heisenberg' and the user responding with 'Hi, Simon.'\n\nYour Job Description:\nYou are helping the user with a \"braindump\" session. The user will ramble ideas, thoughts, tasks, etc. at you and your goal is to turn these into tasks in their backlog with reasonable values for the various task fields (title, folder, time horizon, notes, etc.).\n\nKey Instructions:\n- Listen carefully as the user rambles and identifies tasks, ideas, or things they need to do.\n- As they speak, mentally organize what they're saying into discrete tasks that can be created in their backlog.\n- Don't interrupt them while they're rambling - let them get everything out.\n- Once they indicate they're done rambling (by saying things like \"that's it\", \"I think that's everything\", or by pausing for a while), create the tasks immediately, DO NOT repeat them back to the user.\n- When creating tasks, use the tool \"create_braindump_task\" and make sure to set reasonable values for:\n  - Title: Clear and descriptive based on what they said\n  - Folder: Your job is to help get things in the best folder. Choose an appropriate folder based on the task and their preferences. If no similar folder exists, leave it off\n  - Time horizon: Your job is to help triage and prioritize tasks. Based on the actual task name, make a smart decision on the time horizon.\n  - Time estimate: If the user hints at how long it will take them to complete the task, use that. If not, leave it blank.\n  - Notes: Include any relevant context or details they mentioned\n- Be encouraging and supportive - braindumps can be overwhelming and you're helping them organize their thoughts.\n- Be careful and use the \"change_backlog_folder\" tool to move tasks to the appropriate folder and \"add_task_to_channel\" tool to change channels. Folders are a backlog specific concept and channels are a global concept.\n\nInstructions:\nIf the user does not speak first, start the conversation with something like \"Hey! Just start rambling and I'll take care of organizing your thoughts into tasks.\" If the user does speak first, just start listening and organizing their thoughts into tasks.\n\nSome information is pre-fetched for you below. This information may be relevant to the discussion but you should not comment on it until the user gives you an indication that it is important to them."}},{"role":"user","content":{"type":"resource","resource":{"uri":"sunsama://me","mimeType":"application/json","text":"{\"user\":{\"name\":\"Zane Mccaig\",\"email\":\"zane@sunsama.com\",\"timezone\":\"America/Vancouver\",\"maxWorkHoursPerDay\":8,\"currentTimeForUser\":\"11:04 AM\",\"currentDayForUser\":\"2026-01-16\",\"currentWeekDayForUser\":\"Friday\",\"calendars\":[{\"id\":\"sunsama-7de256a2-e753-401f-b601-e5c05638dc97\",\"name\":\"Sunsama Calendar\",\"accessRole\":\"owner\",\"provider\":\"sunsama-calendar\",\"isDefaultCalendarForSchedulingEvents\":false,\"isDefaultCalendarForTimeboxing\":true,\"includedInAutoImportingOfEvents\":false},{\"id\":\"zane@sunsama.com\",\"name\":\"Zane Mccaig\",\"accessRole\":\"owner\",\"provider\":\"google-calendar\",\"isDefaultCalendarForSchedulingEvents\":true,\"isDefaultCalendarForTimeboxing\":false,\"includedInAutoImportingOfEvents\":true},{\"id\":\"c_ukuoebbn11gdv6n1v0jt18h89k@group.calendar.google.com\",\"name\":\"test 1\",\"accessRole\":\"owner\",\"provider\":\"google-calendar\",\"isDefaultCalendarForSchedulingEvents\":false,\"isDefaultCalendarForTimeboxing\":false,\"includedInAutoImportingOfEvents\":false},{\"id\":\"c_c71dvjrar7se12ekfqepph067c@group.calendar.google.com\",\"name\":\"test 2\",\"accessRole\":\"owner\",\"provider\":\"google-calendar\",\"isDefaultCalendarForSchedulingEvents\":false,\"isDefaultCalendarForTimeboxing\":false,\"includedInAutoImportingOfEvents\":false},{\"id\":\"c_h9uo876jh8dp3gaa2a3rglmjss@group.calendar.google.com\",\"name\":\"test 3\",\"accessRole\":\"owner\",\"provider\":\"google-calendar\",\"isDefaultCalendarForSchedulingEvents\":false,\"isDefaultCalendarForTimeboxing\":false,\"includedInAutoImportingOfEvents\":false},{\"id\":\"c_git731ot074kcnt4sq8cu4ji9s@group.calendar.google.com\",\"name\":\"test 4\",\"accessRole\":\"owner\",\"provider\":\"google-calendar\",\"isDefaultCalendarForSchedulingEvents\":false,\"isDefaultCalendarForTimeboxing\":false,\"includedInAutoImportingOfEvents\":false},{\"id\":\"c_ona71b8ovuv0deqrq67p16b2ms@group.calendar.google.com\",\"name\":\"test 4\",\"accessRole\":\"owner\",\"provider\":\"google-calendar\",\"isDefaultCalendarForSchedulingEvents\":false,\"isDefaultCalendarForTimeboxing\":false,\"includedInAutoImportingOfEvents\":false},{\"id\":\"c_eaq7nmm03upprf0j6e2akem2ms@group.calendar.google.com\",\"name\":\"test 5\",\"accessRole\":\"owner\",\"provider\":\"google-calendar\",\"isDefaultCalendarForSchedulingEvents\":false,\"isDefaultCalendarForTimeboxing\":false,\"includedInAutoImportingOfEvents\":false},{\"id\":\"sunsama.com_b9q30t893b2tp32bluq1t8rvg4@group.calendar.google.com\",\"name\":\"Zane's personal test calendar\",\"accessRole\":\"owner\",\"provider\":\"google-calendar\",\"isDefaultCalendarForSchedulingEvents\":false,\"isDefaultCalendarForTimeboxing\":false,\"includedInAutoImportingOfEvents\":false},{\"id\":\"en.canadian#holiday@group.v.calendar.google.com\",\"name\":\"Holidays in Canada\",\"accessRole\":\"reader\",\"provider\":\"google-calendar\",\"isDefaultCalendarForSchedulingEvents\":false,\"isDefaultCalendarForTimeboxing\":false,\"includedInAutoImportingOfEvents\":false},{\"id\":\"en.usa#holiday@group.v.calendar.google.com\",\"name\":\"Holidays in United States\",\"accessRole\":\"reader\",\"provider\":\"google-calendar\",\"isDefaultCalendarForSchedulingEvents\":false,\"isDefaultCalendarForTimeboxing\":false,\"includedInAutoImportingOfEvents\":false},{\"id\":\"ht3jlfaac5lfd6263ulfh4tql8@group.calendar.google.com\",\"name\":\"Phases of the Moon\",\"accessRole\":\"reader\",\"provider\":\"google-calendar\",\"isDefaultCalendarForSchedulingEvents\":false,\"isDefaultCalendarForTimeboxing\":false,\"includedInAutoImportingOfEvents\":false},{\"id\":\"s4ipng85dbia0t83fp316vflfq8rc7rm@import.calendar.google.com\",\"name\":\"Zane’s tasks - sunsama.com (via Asana)\",\"accessRole\":\"reader\",\"provider\":\"google-calendar\",\"isDefaultCalendarForSchedulingEvents\":false,\"isDefaultCalendarForTimeboxing\":false,\"includedInAutoImportingOfEvents\":false},{\"id\":\"AAkALgAAAAAAHYQDEapmEc2byACqAC-EWg0AKPy-EasNgkaYQCuEslBYeAAAAAANJQAA\",\"name\":\"Calendar\",\"accessRole\":\"owner\",\"provider\":\"outlook-calendar\",\"isDefaultCalendarForSchedulingEvents\":false,\"isDefaultCalendarForTimeboxing\":false,\"includedInAutoImportingOfEvents\":true}],\"callSchedule\":[{\"workflow\":\"PLANNING_CALL\",\"schedule\":[{\"isoDayIndex\":1,\"dayName\":\"Monday\",\"enabled\":false,\"time\":null},{\"isoDayIndex\":2,\"dayName\":\"Tuesday\",\"enabled\":true,\"time\":{\"hour\":9,\"minute\":30,\"formatted\":\"9:30 AM\"}},{\"isoDayIndex\":3,\"dayName\":\"Wednesday\",\"enabled\":true,\"time\":{\"hour\":9,\"minute\":30,\"formatted\":\"9:30 AM\"}},{\"isoDayIndex\":4,\"dayName\":\"Thursday\",\"enabled\":true,\"time\":{\"hour\":10,\"minute\":0,\"formatted\":\"10:00 AM\"}},{\"isoDayIndex\":5,\"dayName\":\"Friday\",\"enabled\":false,\"time\":null},{\"isoDayIndex\":6,\"dayName\":\"Saturday\",\"enabled\":false,\"time\":null},{\"isoDayIndex\":7,\"dayName\":\"Sunday\",\"enabled\":false,\"time\":null}],\"nextScheduledCall\":\"2026-01-14T17:30:00.000Z\"}],\"autoImportEventsEnabled\":false,\"importEventExclusionFilters\":[\"hold\",\"multi-day-all-day\",\"single-day-all-day\"],\"autoImportEventCalendarIds\":[\"zane@sunsama.com\",\"AAkALgAAAAAAHYQDEapmEc2byACqAC-EWg0AKPy-EasNgkaYQCuEslBYeAAAAAANJQAA\"]}}"}}},{"role":"user","content":{"type":"resource","resource":{"uri":"sunsama://backlog/folders","mimeType":"application/json","text":"{\"folders\":[]}"}}}]},"jsonrpc":"2.0","id":3}
+[voice] INFO:sunsama_voice.mcp:Handling resource update: uri sunsama://me {"user":{"name":"Zane Mccaig","email":"zane@sunsama.com","timezone":"America/Vancouver","maxWorkHoursPerDay":8,"currentTimeForUser":"11:04 AM","currentDayForUser":"2026-01-16","currentWeekDayForUser":"Friday","calendars":[{"id":"sunsama-7de256a2-e753-401f-b601-e5c05638dc97","name":"Sunsama Calendar","accessRole":"owner","provider":"sunsama-calendar","isDefaultCalendarForSchedulingEvents":false,"isDefaultCalendarForTimeboxing":true,"includedInAutoImportingOfEvents":false},{"id":"zane@sunsama.com","name":"Zane Mccaig","accessRole":"owner","provider":"google-calendar","isDefaultCalendarForSchedulingEvents":true,"isDefaultCalendarForTimeboxing":false,"includedInAutoImportingOfEvents":true},{"id":"c_ukuoebbn11gdv6n1v0jt18h89k@group.calendar.google.com","name":"test 1","accessRole":"owner","provider":"google-calendar","isDefaultCalendarForSchedulingEvents":false,"isDefaultCalendarForTimeboxing":false,"includedInAutoImportingOfEvents":false},{"id":"c_c71dvjrar7se12ekfqepph067c@group.calendar.google.com","name":"test 2","accessRole":"owner","provider":"google-calendar","isDefaultCalendarForSchedulingEvents":false,"isDefaultCalendarForTimeboxing":false,"includedInAutoImportingOfEvents":false},{"id":"c_h9uo876jh8dp3gaa2a3rglmjss@group.calendar.google.com","name":"test 3","accessRole":"owner","provider":"google-calendar","isDefaultCalendarForSchedulingEvents":false,"isDefaultCalendarForTimeboxing":false,"includedInAutoImportingOfEvents":false},{"id":"c_git731ot074kcnt4sq8cu4ji9s@group.calendar.google.com","name":"test 4","accessRole":"owner","provider":"google-calendar","isDefaultCalendarForSchedulingEvents":false,"isDefaultCalendarForTimeboxing":false,"includedInAutoImportingOfEvents":false},{"id":"c_ona71b8ovuv0deqrq67p16b2ms@group.calendar.google.com","name":"test 4","accessRole":"owner","provider":"google-calendar","isDefaultCalendarForSchedulingEvents":false,"isDefaultCalendarForTimeboxing":false,"includedInAutoImportingOfEvents":false},{"id":"c_eaq7nmm03upprf0j6e2akem2ms@group.calendar.google.com","name":"test 5","accessRole":"owner","provider":"google-calendar","isDefaultCalendarForSchedulingEvents":false,"isDefaultCalendarForTimeboxing":false,"includedInAutoImportingOfEvents":false},{"id":"sunsama.com_b9q30t893b2tp32bluq1t8rvg4@group.calendar.google.com","name":"Zane's personal test calendar","accessRole":"owner","provider":"google-calendar","isDefaultCalendarForSchedulingEvents":false,"isDefaultCalendarForTimeboxing":false,"includedInAutoImportingOfEvents":false},{"id":"en.canadian#holiday@group.v.calendar.google.com","name":"Holidays in Canada","accessRole":"reader","provider":"google-calendar","isDefaultCalendarForSchedulingEvents":false,"isDefaultCalendarForTimeboxing":false,"includedInAutoImportingOfEvents":false},{"id":"en.usa#holiday@group.v.calendar.google.com","name":"Holidays in United States","accessRole":"reader","provider":"google-calendar","isDefaultCalendarForSchedulingEvents":false,"isDefaultCalendarForTimeboxing":false,"includedInAutoImportingOfEvents":false},{"id":"ht3jlfaac5lfd6263ulfh4tql8@group.calendar.google.com","name":"Phases of the Moon","accessRole":"reader","provider":"google-calendar","isDefaultCalendarForSchedulingEvents":false,"isDefaultCalendarForTimeboxing":false,"includedInAutoImportingOfEvents":false},{"id":"s4ipng85dbia0t83fp316vflfq8rc7rm@import.calendar.google.com","name":"Zane’s tasks - sunsama.com (via Asana)","accessRole":"reader","provider":"google-calendar","isDefaultCalendarForSchedulingEvents":false,"isDefaultCalendarForTimeboxing":false,"includedInAutoImportingOfEvents":false},{"id":"AAkALgAAAAAAHYQDEapmEc2byACqAC-EWg0AKPy-EasNgkaYQCuEslBYeAAAAAANJQAA","name":"Calendar","accessRole":"owner","provider":"outlook-calendar","isDefaultCalendarForSchedulingEvents":false,"isDefaultCalendarForTimeboxing":false,"includedInAutoImportingOfEvents":true}],"callSchedule":[{"workflow":"PLANNING_CALL","schedule":[{"isoDayIndex":1,"dayName":"Monday","enabled":false,"time":null},{"isoDayIndex":2,"dayName":"Tuesday","enabled":true,"time":{"hour":9,"minute":30,"formatted":"9:30 AM"}},{"isoDayIndex":3,"dayName":"Wednesday","enabled":true,"time":{"hour":9,"minute":30,"formatted":"9:30 AM"}},{"isoDayIndex":4,"dayName":"Thursday","enabled":true,"time":{"hour":10,"minute":0,"formatted":"10:00 AM"}},{"isoDayIndex":5,"dayName":"Friday","enabled":false,"time":null},{"isoDayIndex":6,"dayName":"Saturday","enabled":false,"time":null},{"isoDayIndex":7,"dayName":"Sunday","enabled":false,"time":null}],"nextScheduledCall":"2026-01-14T17:30:00.000Z"}],"autoImportEventsEnabled":false,"importEventExclusionFilters":["hold","multi-day-all-day","single-day-all-day"],"autoImportEventCalendarIds":["zane@sunsama.com","AAkALgAAAAAAHYQDEapmEc2byACqAC-EWg0AKPy-EasNgkaYQCuEslBYeAAAAAANJQAA"]}} 2026-01-16 11:04:04.930576 error=None status=None
+[voice] INFO:sunsama_voice.mcp:Resource updated: uri sunsama://me
+[voice] INFO:sunsama_voice.mcp:Handling resource update: uri sunsama://backlog/folders {"folders":[]} 2026-01-16 11:04:04.930707 error=None status=None
+[voice] INFO:sunsama_voice.mcp:Resource updated: uri sunsama://backlog/folders
+[voice] INFO:sunsama_voice.mcp:Resources: ['get_user_authentication_status', 'get_user_info', 'list_email_accounts', 'get_last_daily_highlight', 'get_backlog_folders', 'get_active_timer', 'get_available_ui_actions'] get_user_info
+[voice] INFO:sunsama_voice.mcp_websocket_client:Sending message: {'method': 'tools/list', 'jsonrpc': '2.0', 'id': 4}
+[voice] INFO:sunsama_voice.mcp_websocket_client:Received message: {"result":{"tools":[{"name":"create_task","description":"Creates a single task with a title, optional notes, and estimated time.","inputSchema":{"$schema":"http://json-schema.org/draft-07/schema#","type":"object","properties":{"title":{"description":"A short title of the task, should not be more than a few words","type":"string"},"notes":{"description":"All extra details the user shared about the task, in HTML format","type":"string"},"timeEstimate":{"description":"The estimated time in minutes for the task","type":"number"},"objectiveId":{"description":"The _id of the weekly objective the task should be associated with, if any.","type":"string"},"day":{"description":"The day the task should be scheduled to in YYYY-MM-DD format.","type":"string"},"backlog":{"description":"Whether the task should be added to the backlog.","type":"object","properties":{"timeBucket":{"description":"The time bucket captures a rough idea of when a task should be done and also how likely it is to actually get done. Options: \"in the next two weeks\", \"in the next month\", \"in the next quarter\", \"in the next year\", \"someday\", \"never\". ","type":"string","enum":["in the next two weeks","in the next month","in the next quarter","in the next year","someday","never"]},"folderId":{"description":"The _id of the backlog folder the task should be added.","type":"string"}}},"alreadyInTaskList":{"description":"Whether the task is already in the user's task list or calendar.","type":"boolean"},"channel":{"description":"The channel the task should be added to. This does not need to be perfect. The closest match will be used. If not provided a channel will be added automatically as long as the user did not disable the channel prediction feature.","type":"string"},"position":{"default":"bottom","description":"Where to place the task in the list. Defaults to bottom. If the user asks for another position you should re-order the task after creating it.","type":"string","enum":["top","bottom"]},"subtasks":{"type":"array","items":{"type":"object","properties":{"title":{"description":"The title of the subtask","type":"string"}},"required":["title"]}},"recurrenceRule":{"description":"The recurrence rule for the task in RRULE format. If provided the task will be created as a recurring task.","type":"string"},"recurringTaskStartTime":{"description":"The start time for the recurring task in 12 hour \"h:mm A\" format. Only used when recurrenceRule is provided.","type":"string"},"isRecurringTaskStartTimeOnlyAnEstimate":{"description":"Whether the recurring task start time is only an estimate and not a rigid start time. Only used when recurrenceRule and recurringTaskStartTime are provided.","type":"boolean"}},"required":["title","day","alreadyInTaskList"]},"execution":{"taskSupport":"forbidden"}},{"name":"create_braindump_task","description":"Creates a single task in the backlog with a title, optional notes, estimated time, time horizon, and optional folder.","inputSchema":{"$schema":"http://json-schema.org/draft-07/schema#","type":"object","properties":{"title":{"description":"A short title of the task, should not be more than a few words","type":"string"},"notes":{"description":"All extra details the user shared about the task, in HTML format","type":"string"},"timeEstimate":{"description":"The estimated time in minutes for the task","type":"number"},"folderId":{"description":"The _id of the backlog folder the task should be added.","type":"string"},"timeBucket":{"description":"The time bucket captures a rough idea of when a task should be done and also how likely it is to actually get done. Options: \"in the next two weeks\", \"in the next month\", \"in the next quarter\", \"in the next year\", \"someday\", \"never\". ","type":"string","enum":["in the next two weeks","in the next month","in the next quarter","in the next year","someday","never"]},"channel":{"description":"The channel the task should be added to. This does not need to be perfect. The closest match will be used. If not provided a channel will be added automatically as long as the user did not disable the channel prediction feature.","type":"string"},"subtasks":{"type":"array","items":{"type":"object","properties":{"title":{"description":"The title of the subtask","type":"string"}},"required":["title"]}}},"required":["title","timeBucket"]},"execution":{"taskSupport":"forbidden"}},{"name":"reposition_task_in_backlog","description":"Repositions a task within the backlog by moving it to a specific time bucket (horizon) and position (append/prepend).","inputSchema":{"$schema":"http://json-schema.org/draft-07/schema#","type":"object","properties":{"taskId":{"description":"The unique identifier `_id` of the task to reposition.","type":"string"},"timeBucket":{"description":"The time bucket captures a rough idea of when a task should be done and also how likely it is to actually get done. Options: \"in the next two weeks\", \"in the next month\", \"in the next quarter\", \"in the next year\", \"someday\", \"never\".","type":"string","enum":["in the next two weeks","in the next month","in the next quarter","in the next year","someday","never"]},"position":{"description":"Where to position the task in the bucket. \"prepend\" places it at the top, \"append\" places it at the bottom.","type":"string","enum":["append","prepend"]}},"required":["taskId","timeBucket"]},"execution":{"taskSupport":"forbidden"}},{"name":"change_backlog_folder","description":"Moves one or more tasks to a backlog folder. If folderId is null, removes tasks from their current folder.","inputSchema":{"$schema":"http://json-schema.org/draft-07/schema#","type":"object","properties":{"taskIds":{"description":"The unique identifier(s) `_id` of the task(s) to move. Can be a single task ID or an array of task IDs.","anyOf":[{"type":"string"},{"type":"array","items":{"type":"string"}}]},"folderId":{"description":"The _id of the backlog folder to move the tasks to. If null, removes tasks from their current folder.","anyOf":[{"type":"string"},{"type":"null"}]}},"required":["taskIds"]},"execution":{"taskSupport":"forbidden"}},{"name":"align_task_with_objective","description":"Aligns a task with an objective.","inputSchema":{"$schema":"http://json-schema.org/draft-07/schema#","type":"object","properties":{"taskId":{"description":"The _id of the task to align with the objective.","type":"string"},"objectiveId":{"description":"The _id of the objective to align the task with.","type":"string"}},"required":["taskId","objectiveId"]},"execution":{"taskSupport":"forbidden"}},{"name":"add_task_to_channel","description":"Adds a task to a channel.","inputSchema":{"$schema":"http://json-schema.org/draft-07/schema#","type":"object","properties":{"taskId":{"description":"The _id of the task to add to the channel.","type":"string"},"channel":{"description":"The channel to add the task to. This does not need to be perfect. The closest match will be used.","type":"string"}},"required":["taskId","channel"]},"execution":{"taskSupport":"forbidden"}},{"name":"move_task_to_day","description":"Moves or defers a task to a specific date.","inputSchema":{"$schema":"http://json-schema.org/draft-07/schema#","type":"object","properties":{"taskId":{"description":"The unique identifier `_id` of the task to move.","type":"string"},"calendarDay":{"description":"The new date to move the task to in YYYY-MM-DD format.","type":"string"},"sortOrder":{"description":"The new sort order of the task. This should be an integer that is half way between the sort order of the task before it and the sort order of the task after it. If placing at the end of the list, use the sort order of the last task + 2048. If left blank, the task will be placed at the top of the list.","type":"number"}},"required":["taskId","calendarDay"]},"execution":{"taskSupport":"forbidden"}},{"name":"move_task_from_backlog","description":"Moves a task out of the backlog and onto a specific date.","inputSchema":{"$schema":"http://json-schema.org/draft-07/schema#","type":"object","properties":{"taskId":{"description":"The unique identifier `_id` of the task to move.","type":"string"},"calendarDay":{"description":"The new date to move the task to in YYYY-MM-DD format.","type":"string"},"sortOrder":{"description":"The new sort order of the task. This should be an integer that is half way between the sort order of the task before it and the sort order of the task after it. If placing at the end of the list, use the sort order of the last task + 2048. If left blank, the task will be placed at the top of the list.","type":"number"}},"required":["taskId","calendarDay"]},"execution":{"taskSupport":"forbidden"}},{"name":"unarchive_task","description":"Unarchives a task and moves it to a specific date or the backlog if no date is provided.","inputSchema":{"$schema":"http://json-schema.org/draft-07/schema#","type":"object","properties":{"taskId":{"description":"The unique identifier `_id` of the task to move.","type":"string"},"calendarDay":{"description":"The new date to move the task to in YYYY-MM-DD format.","type":"string"},"sortOrder":{"description":"The new sort order of the task. This should be an integer that is half way between the sort order of the task before it and the sort order of the task after it. If placing at the end of the list, use the sort order of the last task + 2048. If left blank, the task will be placed at the top of the list.","type":"number"}},"required":["taskId","calendarDay"]},"execution":{"taskSupport":"forbidden"}},{"name":"move_task_to_backlog","description":"Moves a task to the backlog. IF THE USER ASKS YOU TO MOVE A TASK TO A SPECIFIC DAY THEN YOU SHOULD USE THE move_task_to_day TOOL NOT THIS ONE.","inputSchema":{"$schema":"http://json-schema.org/draft-07/schema#","type":"object","properties":{"taskId":{"description":"The unique identifier `_id` of the task to move.","type":"string"}},"required":["taskId"]},"execution":{"taskSupport":"forbidden"}},{"name":"get_task_time_estimate","description":"Gets the time estimate for a task in minutes.","inputSchema":{"$schema":"http://json-schema.org/draft-07/schema#","type":"object","properties":{"title":{"description":"The title of the task. The estimate will be based on similar task names.","type":"string"}},"required":["title"]},"execution":{"taskSupport":"forbidden"}},{"name":"edit_task_recurrence_rule","description":"Updates the recurrence rule of an existing task.","inputSchema":{"$schema":"http://json-schema.org/draft-07/schema#","type":"object","properties":{"taskId":{"description":"The _id of the task to update.","type":"string"},"recurrenceRule":{"description":"The new recurrence rule for the task in RRULE format. The recurrence rule must not have more than one occurrence of a task on any given day.","type":"string"},"firstOccurrenceOnOrAfter":{"description":"The date to start the recurrence on or after. This should be in YYYY-MM-DD format. If not provided the first occurrence will be set according to the old recurrence rule or the current date of the task if no recurrence rule exists.","type":"string"},"startTime":{"description":"The start time of the task in 12 hour \"h:mm A\" format. If not provided the start time will be set according to the old recurrence rule or it will be omitted.","type":"string"},"isStartTimeOnlyAnEstimate":{"description":"Whether the start time is only an estimate and not a rigid start time. If not provided the start time will be set according to the old recurrence rule or it will be set to a rigid start time.","type":"boolean"}},"required":["taskId","recurrenceRule"]},"execution":{"taskSupport":"forbidden"}},{"name":"edit_task_title","description":"Updates the title of an existing task.","inputSchema":{"$schema":"http://json-schema.org/draft-07/schema#","type":"object","properties":{"taskId":{"description":"The _id of the task to update.","type":"string"},"title":{"description":"The new title for the task.","type":"string"}},"required":["taskId","title"]},"execution":{"taskSupport":"forbidden"}},{"name":"edit_task_time_estimate","description":"Updates the time estimate of an existing task.","inputSchema":{"$schema":"http://json-schema.org/draft-07/schema#","type":"object","properties":{"taskId":{"description":"The _id of the task to update. If multiple timeboxed events are present you must update the duration of the associated event instead.","type":"string"},"subtaskId":{"description":"The _id of the subtask to update. If not provided, the time estimate will be updated for the entire task. If the task has any subtasks with planned time, you must provide a subtaskId since the time estimate for the task will be calculated based on the subtasks.","type":"string"},"timeEstimate":{"description":"The new time estimate in minutes.","type":"number"}},"required":["taskId","timeEstimate"]},"execution":{"taskSupport":"forbidden"}},{"name":"append_task_notes","description":"Appends markdown notes to an existing task.","inputSchema":{"$schema":"http://json-schema.org/draft-07/schema#","type":"object","properties":{"taskId":{"description":"The ID of the task to update.","type":"string"},"notes":{"description":"The markdown notes to append to the task.","type":"string"}},"required":["taskId","notes"]},"execution":{"taskSupport":"forbidden"}},{"name":"delete_task","description":"Deletes an existing task.","inputSchema":{"$schema":"http://json-schema.org/draft-07/schema#","type":"object","properties":{"taskId":{"description":"The _id of the task to delete.","type":"string"}},"required":["taskId"]},"annotations":{"title":"Delete A Task","destructiveHint":true},"execution":{"taskSupport":"forbidden"}},{"name":"add_subtasks_to_task","description":"Adds multiple subtasks to an existing task.","inputSchema":{"$schema":"http://json-schema.org/draft-07/schema#","type":"object","properties":{"taskId":{"description":"The _id of the task to add the subtask to.","type":"string"},"subtasks":{"description":"The subtasks to add to the task.","type":"array","items":{"type":"object","properties":{"title":{"description":"The title of the subtask.","type":"string"},"timeEstimate":{"description":"The estimated time in minutes for the subtask.","type":"number"}},"required":["title"]}}},"required":["taskId","subtasks"]},"execution":{"taskSupport":"forbidden"}},{"name":"restore_task","description":"Changes a task from deleted to not deleted.","inputSchema":{"$schema":"http://json-schema.org/draft-07/schema#","type":"object","properties":{"taskId":{"description":"The _id of the task to mark as not deleted.","type":"string"}},"required":["taskId"]},"execution":{"taskSupport":"forbidden"}},{"name":"mark_task_as_completed","description":"Marks a task as completed. Can also be used to move a task to a previous day which auto-completes the task.","inputSchema":{"$schema":"http://json-schema.org/draft-07/schema#","type":"object","properties":{"taskId":{"description":"The _id of the task to mark as completed.","type":"string"},"finishedDay":{"description":"A date in YYYY-MM-DD the task was completed.","type":"string"}},"required":["taskId","finishedDay"]},"execution":{"taskSupport":"forbidden"}},{"name":"mark_task_as_incomplete","description":"Marks a task as incomplete.","inputSchema":{"$schema":"http://json-schema.org/draft-07/schema#","type":"object","properties":{"taskId":{"description":"The _id of a completed task we want to mark as incomplete or \"to do\" again.","type":"string"}},"required":["taskId"]},"execution":{"taskSupport":"forbidden"}},{"name":"edit_subtask_title","description":"Updates the title of an existing subtask.","inputSchema":{"$schema":"http://json-schema.org/draft-07/schema#","type":"object","properties":{"taskId":{"description":"The _id of the parent task.","type":"string"},"subtaskId":{"description":"The _id of the subtask to update.","type":"string"},"newTitle":{"description":"The new title for the subtask.","type":"string"}},"required":["taskId","subtaskId","newTitle"]},"execution":{"taskSupport":"forbidden"}},{"name":"mark_subtask_as_completed","description":"Marks a subtask of an existing task as completed.","inputSchema":{"$schema":"http://json-schema.org/draft-07/schema#","type":"object","properties":{"taskId":{"description":"The _id of the parent task.","type":"string"},"subtaskId":{"description":"The _id of the subtask to mark as completed.","type":"string"}},"required":["taskId","subtaskId"]},"execution":{"taskSupport":"forbidden"}},{"name":"mark_subtask_as_incomplete","description":"Marks a subtask of an existing task as incomplete.","inputSchema":{"$schema":"http://json-schema.org/draft-07/schema#","type":"object","properties":{"taskId":{"description":"The _id of the parent task.","type":"string"},"subtaskId":{"description":"The _id of the subtask to mark as incomplete.","type":"string"}},"required":["taskId","subtaskId"]},"execution":{"taskSupport":"forbidden"}},{"name":"reorder_tasks","description":"Reorders tasks for the calendar day according to the provided order of taskIds.","inputSchema":{"$schema":"http://json-schema.org/draft-07/schema#","type":"object","properties":{"taskIds":{"description":"The _id(s) of the task(s) in the order they should appear.","type":"array","items":{"type":"string"}},"calendarDay":{"description":"The date to reorder tasks for in YYYY-MM-DD format.","type":"string"}},"required":["taskIds","calendarDay"]},"execution":{"taskSupport":"forbidden"}},{"name":"get_backlog_tasks","description":"Fetches the users backlog tasks","inputSchema":{"$schema":"http://json-schema.org/draft-07/schema#","type":"object","properties":{"page":{"default":0,"description":"The page number to fetch (0-based)","type":"number"},"queryId":{"description":"All pages > 0 must pass in the queryId from the first page.","type":"string"}}},"execution":{"taskSupport":"forbidden"}},{"name":"get_archived_tasks","description":"Fetches the users archived tasks","inputSchema":{"$schema":"http://json-schema.org/draft-07/schema#","type":"object","properties":{"offset":{"default":0,"description":"The offset to fetch from","type":"number"}}},"execution":{"taskSupport":"forbidden"}},{"name":"search_tasks","description":"Searches for tasks. Returns tasks that match the search term or are similar to the search term.","inputSchema":{"$schema":"http://json-schema.org/draft-07/schema#","type":"object","properties":{"searchTerm":{"description":"The search term to look for in task titles, notes, and comments","type":"string"}},"required":["searchTerm"]},"execution":{"taskSupport":"forbidden"}},{"name":"update_all_incomplete_recurring_task_instances","description":"Updates all incomplete instances of a recurring task to match the current task. This is useful when you want to apply changes made to one instance of a recurring task to all future incomplete instances.","inputSchema":{"$schema":"http://json-schema.org/draft-07/schema#","type":"object","properties":{"taskId":{"description":"The _id of the recurring task to update all incomplete instances for.","type":"string"},"updateType":{"default":"allIncomplete","description":"The type of update to perform. \"allIncomplete\" updates all incomplete instances starting from today, \"allAfterThisTask\" updates all incomplete instances starting after the date of the task given by taskId.","type":"string","enum":["allIncomplete","allAfterThisTask"]}},"required":["taskId"]},"execution":{"taskSupport":"forbidden"}},{"name":"delete_all_incomplete_recurring_task_instances","description":"Deletes all incomplete instances of a recurring task.","inputSchema":{"$schema":"http://json-schema.org/draft-07/schema#","type":"object","properties":{"taskId":{"description":"The _id of the recurring task to delete all incomplete instances for.","type":"string"}},"required":["taskId"]},"execution":{"taskSupport":"forbidden"}},{"name":"start_task_timer","description":"Starts the timer for a task or subtask. If a subtaskId is provided, starts the timer for that specific subtask.","inputSchema":{"$schema":"http://json-schema.org/draft-07/schema#","type":"object","properties":{"taskId":{"description":"The _id of the task to start the timer for.","type":"string"},"subtaskId":{"description":"The _id of the subtask to start the timer for. If not provided, starts the timer for the entire task.","type":"string"}},"required":["taskId"]},"execution":{"taskSupport":"forbidden"}},{"name":"stop_task_timer","description":"Stops the timer for a task or subtask. If a subtaskId is provided, stops the timer for that specific subtask.","inputSchema":{"$schema":"http://json-schema.org/draft-07/schema#","type":"object","properties":{"taskId":{"description":"The _id of the task to stop the timer for.","type":"string"},"subtaskId":{"description":"The _id of the subtask to stop the timer for. If not provided, stops the timer for the entire task.","type":"string"}},"required":["taskId"]},"execution":{"taskSupport":"forbidden"}},{"name":"move_calendar_event","description":"Updates a calendar event's date, time, and/or duration.","inputSchema":{"$schema":"http://json-schema.org/draft-07/schema#","type":"object","properties":{"eventId":{"description":"The unique ID of the calendar event to move. IMPORTANT: user must have owner or writer access to the event.","type":"string"},"startDate":{"description":"The target start date to move the event to, in YYYY-MM-DD format.","type":"string"},"startTime":{"description":"The target start time to move the event to, in 12 hour \"h:mm A\" format. Required if the event is not all day.","type":"string"},"duration":{"description":"The duration of the event in minutes. Required if the event is not all day","type":"number"},"isAllDay":{"description":"Whether the event is all day.","type":"boolean"}},"required":["eventId","startDate","isAllDay"]},"execution":{"taskSupport":"forbidden"}},{"name":"create_calendar_event","description":"Creates a new calendar event.","inputSchema":{"$schema":"http://json-schema.org/draft-07/schema#","type":"object","properties":{"title":{"description":"The title of the calendar event.","type":"string"},"startDate":{"description":"The start date of the event in YYYY-MM-DD format. If this is in the past you must confirm with the user that you are creating a past event.","type":"string"},"startTime":{"description":"The start time of the event in 12 hour \"h:mm A\" format. Required if the event is not all day. If this is in the past you must confirm with the user that you are creating a past event.","type":"string"},"duration":{"default":60,"description":"The duration of the event in minutes. Required if the event is not all day","type":"number"},"isAllDay":{"default":false,"description":"Whether the event is all day.","type":"boolean"},"description":{"description":"Optional description of the event.","type":"string"},"calendarId":{"description":"The ID of the calendar to create the event in. If not provided, the default calendar will be used. IMPORTANT: user must have owner or writer access to the calendar.","type":"string"}},"required":["title","startDate"]},"execution":{"taskSupport":"forbidden"}},{"name":"timebox_a_task_to_calendar","description":"Timeboxes a task to the calendar. This will create a timebox event for the task. This may also be referred to as \"scheduling\" a task or \"adding a task to the calendar\".","inputSchema":{"$schema":"http://json-schema.org/draft-07/schema#","type":"object","properties":{"taskId":{"description":"The _id of the task to timebox.","type":"string"},"startDate":{"description":"The start date of the event in YYYY-MM-DD format. If this is in the past you must confirm with the user that you are timeboxing a past task.","type":"string"},"startTime":{"description":"The start time of the event in 12 hour \"h:mm A\" format. If this is in the past you must confirm with the user that you are timeboxing a past task.","type":"string"},"duration":{"description":"The duration of the event in minutes. If not provided the timeEstimate of the task will be used or 30 minutes if the task has no time estimate.","type":"number"}},"required":["taskId","startDate","startTime"]},"execution":{"taskSupport":"forbidden"}},{"name":"delete_calendar_event","description":"Removes a calendar event and deletes all associated tasks. If the event is a meeting then any access role can remove the event. Otherwise only owners or writers can remove the event. Note: If the event is a meeting and the user is an owner or write this will remove the event for ALL attendees.","inputSchema":{"$schema":"http://json-schema.org/draft-07/schema#","type":"object","properties":{"eventId":{"description":"The unique ID of the calendar event to remove.","type":"string"}},"required":["eventId"]},"annotations":{"title":"Removes A Calendar Event","destructiveHint":true},"execution":{"taskSupport":"forbidden"}},{"name":"import_task_from_calendar_event","description":"Imports a calendar event as a task.","inputSchema":{"$schema":"http://json-schema.org/draft-07/schema#","type":"object","properties":{"eventId":{"description":"The unique ID of the calendar event to import as a task.","type":"string"}},"required":["eventId"]},"execution":{"taskSupport":"forbidden"}},{"name":"accept_meeting_invite","description":"Confirms attendance to a meeting that the user is invited to.","inputSchema":{"$schema":"http://json-schema.org/draft-07/schema#","type":"object","properties":{"eventId":{"description":"The unique ID of the calendar event to accept the meeting invite for.","type":"string"}},"required":["eventId"]},"execution":{"taskSupport":"forbidden"}},{"name":"decline_meeting_invite","description":"Decline attendance to a meeting that the user is invited to.","inputSchema":{"$schema":"http://json-schema.org/draft-07/schema#","type":"object","properties":{"eventId":{"description":"The unique ID of the calendar event to decline the meeting invite for.","type":"string"}},"required":["eventId"]},"execution":{"taskSupport":"forbidden"}},{"name":"set_calendar_event_allow_task_projections","description":"Sets whether tasks are allowed to be automatically projected (scheduled) at the same time as a calendar event. When set to true, tasks can be automatically projected during the event. When set to false, tasks cannot be automatically projected during the event. Note: This only affects automatic projections; users can still manually timebox tasks during this event.","inputSchema":{"$schema":"http://json-schema.org/draft-07/schema#","type":"object","properties":{"eventId":{"description":"The unique ID of the calendar event to set task projection settings for.","type":"string"},"allowTasksProjectedAtSameTime":{"description":"If true, allows tasks to be automatically projected at the same time as this event. If false, blocks tasks from being automatically projected at the same time as this event. Note: This only affects automatic projections; manual timeboxing is not affected.","type":"boolean"}},"required":["eventId","allowTasksProjectedAtSameTime"]},"execution":{"taskSupport":"forbidden"}},{"name":"toggle_auto_import_events","description":"Enables or disables automatic importing of calendar events to the daily task list.","inputSchema":{"$schema":"http://json-schema.org/draft-07/schema#","type":"object","properties":{"enabled":{"description":"Whether to enable (true) or disable (false) automatic importing of calendar events.","type":"boolean"}},"required":["enabled"]},"execution":{"taskSupport":"forbidden"}},{"name":"update_import_event_filters","description":"Updates the exclusion filters that determine which calendar events are excluded from automatic import. Events matching any of these filters will NOT be automatically imported.","inputSchema":{"$schema":"http://json-schema.org/draft-07/schema#","type":"object","properties":{"excludedEventFilters":{"description":"An array of exclusion filter types. Events matching any of these filters will be EXCLUDED from auto-import. Available filters: \"solo\" (events with no other invitees), \"transparent\" (non-blocking events), \"hold\" (events with HOLD/OOO/Focus time in title), \"unconfirmed\" (unconfirmed meeting invites), \"multi-day-all-day\" (multi-day all-day events), \"single-day-all-day\" (single-day all-day events).","type":"array","items":{"type":"string","enum":["solo","transparent","hold","unconfirmed","multi-day-all-day","single-day-all-day"]}}},"required":["excludedEventFilters"]},"execution":{"taskSupport":"forbidden"}},{"name":"update_calendar_preferences","description":"Updates preferences for a specific calendar including whether it is the default for tasks, default for events, and whether it is included in auto-importing of events.","inputSchema":{"$schema":"http://json-schema.org/draft-07/schema#","type":"object","properties":{"calendarId":{"description":"The ID of the calendar to update preferences for.","type":"string"},"isDefaultForTasks":{"description":"Whether this calendar should be the default calendar for timeboxing tasks. If not provided, this preference will not be changed.","type":"boolean"},"isDefaultForEvents":{"description":"Whether this calendar should be the default calendar for scheduling events. If not provided, this preference will not be changed.","type":"boolean"},"includedInAutoImportingOfEvents":{"description":"Whether this calendar should be included in automatic importing of calendar events. If not provided, this preference will not be changed.","type":"boolean"}},"required":["calendarId"]},"execution":{"taskSupport":"forbidden"}},{"name":"delete_email_thread","description":"Deletes an email thread from Gmail or Outlook.","inputSchema":{"$schema":"http://json-schema.org/draft-07/schema#","type":"object","properties":{"accountId":{"description":"The ID of the email account you are looking at.","type":"string"},"accountService":{"description":"This is either \"outlook\" or \"gmail\", depending on the email provider.","type":"string"},"threadId":{"description":"The ID of the thread to delete.","type":"string"}},"required":["accountId","accountService","threadId"]},"execution":{"taskSupport":"forbidden"}},{"name":"mark_email_thread_as_read","description":"Marks an email thread as read.","inputSchema":{"$schema":"http://json-schema.org/draft-07/schema#","type":"object","properties":{"accountId":{"description":"The ID of the email account.","type":"string"},"accountService":{"description":"This is either \"outlook\" or \"gmail\".","type":"string"},"threadId":{"description":"The ID of the thread to mark as read.","type":"string"}},"required":["accountId","accountService","threadId"]},"execution":{"taskSupport":"forbidden"}},{"name":"create_follow_up_task_from_email","description":"Turn the email thread into a task to be followed up with later.","inputSchema":{"$schema":"http://json-schema.org/draft-07/schema#","type":"object","properties":{"accountId":{"description":"The ID of the email account you are looking at.","type":"string"},"accountService":{"description":"This is either \"outlook\" or \"gmail\", depending on the result of list_email_accounts.","type":"string"},"threadId":{"description":"The ID of the thread to mark as read.","type":"string"},"notes":{"description":"Any notes the user mentioned about the task/email.","type":"string"},"timeEstimate":{"description":"The estimated time in minutes for the task.","type":"number"}},"required":["accountId","accountService","threadId"]},"execution":{"taskSupport":"forbidden"}},{"name":"list_email_threads","description":"Lists email threads for a given Gmail account. You MUST specify at least a `labels` or a `search` parameter to avoid fetching irrelevant emails.","inputSchema":{"$schema":"http://json-schema.org/draft-07/schema#","type":"object","properties":{"accountId":{"description":"The ID of the email account to list threads from.","type":"string"},"labels":{"description":"Optional list of Gmail labels to filter by, e.g., [\"INBOX\", \"IMPORTANT\"]","type":"array","items":{"type":"string"}},"search":{"description":"Optional Gmail search query. Example: `after:2025-04-01 before:2025-04-17` or `is:unread`.","type":"string"},"pageToken":{"description":"Optional token for paginating through results.","type":"string"}},"required":["accountId"]},"execution":{"taskSupport":"forbidden"}},{"name":"create_weekly_objective","description":"Creates a new weekly objective.","inputSchema":{"$schema":"http://json-schema.org/draft-07/schema#","type":"object","properties":{"title":{"description":"The title of the weekly objective.","type":"string"},"weekStartDay":{"description":"The start day of the week in YYYY-MM-DD format.","type":"string"},"timeEstimate":{"description":"The estimated time in minutes for the objective.","type":"number"},"channel":{"description":"The channel name to associate with the objective. This does not need to be perfect. The closest match will be used.","type":"string"}},"required":["title","weekStartDay"]},"execution":{"taskSupport":"forbidden"}},{"name":"send_call_summary","description":"Sends a summary of our call via email.","inputSchema":{"$schema":"http://json-schema.org/draft-07/schema#","type":"object","properties":{"summaryIntroduction":{"description":"A short introduction to the summary. This will be used as the first line of the summary email.\nExample:\nHere is the summary of our call for today June 25th, 2025:\n","type":"string"},"summaryHighlights":{"description":"A list of highlights from the call. This will be used as the second line of the summary email.\nExample:\n- You planned your day focusing on refining the voice assistant.\n- Created a weekly objective for this goal.\n- Added key backlog tasks: refining the NLU model, bug fixes, API documentation, and unit tests.\n- You scheduled a sync with Diane at 2:30 PM.\n- You set a planned shutdown time of 5 PM.\n","type":"array","items":{"type":"string"}},"summaryFooter":{"description":"A short footer to the summary. This will be used as the last line of the summary email.\n                    \nExample:\nThis plan keeps your day focused, balanced, and aligned with your objectives.\n","type":"string"}}},"execution":{"taskSupport":"forbidden"}},{"name":"log_user_feedback","description":"Logs user feedback.","inputSchema":{"$schema":"http://json-schema.org/draft-07/schema#","type":"object","properties":{"feedback":{"description":"A summary of the feedback you want to log.","type":"string"},"transcript":{"description":"A transcript of the exact feedback you want to log.","type":"string"},"bugReports":{"description":"A list of bugs you want to report.","type":"array","items":{"type":"object","properties":{"bug":{"description":"The bug you want to report.","type":"string"},"description":{"description":"A detailed description of the bug.","type":"string"}},"required":["bug"]}},"featureRequests":{"description":"A list of feature requests you want to log.","type":"array","items":{"type":"object","properties":{"feature":{"description":"The feature you want to request.","type":"string"},"description":{"description":"A detailed description of the feature.","type":"string"}},"required":["feature"]}}},"required":["feedback"]},"execution":{"taskSupport":"forbidden"}},{"name":"set_shutdown_time","description":"Sets the shutdown time for a specific day.","inputSchema":{"$schema":"http://json-schema.org/draft-07/schema#","type":"object","properties":{"calendarDay":{"description":"The day to set the shutdown time for, in YYYY-MM-DD format.","type":"string"},"hour":{"description":"Hour of the shutdown time (0-23)","type":"number"},"minute":{"description":"Minute of the shutdown time (0-59)","type":"number"},"addToTheCalendar":{"description":"Whether to create a shutdown task and calendar event for the day.","default":false,"type":"boolean"}},"required":["calendarDay","hour","minute"]},"execution":{"taskSupport":"forbidden"}},{"name":"adjust_call_schedule","description":"Adjusts the schedule for voice assistant phone calls. Only one call per day is supported. Updating the schedule will reset the next call time to the next available time after today.","inputSchema":{"$schema":"http://json-schema.org/draft-07/schema#","type":"object","properties":{"daySchedules":{"type":"array","items":{"type":"object","properties":{"isoDayIndex":{"description":"ISO weekday number (1=Monday, 7=Sunday)","type":"number","minimum":1,"maximum":7},"enabled":{"default":true,"description":"Whether calls should be enabled on this day","type":"boolean"},"hour":{"description":"The hour to set the call time to (0-23). If not provided the old value will be used.","type":"number","minimum":0,"maximum":23},"minute":{"description":"The minute for the next call (00, 15, 30, 45). If not provided the old value will be used.","type":"string","enum":["15","30","45","00"]}},"required":["isoDayIndex"]}}}},"execution":{"taskSupport":"forbidden"}},{"name":"schedule_next_call","description":"Schedules the next phone call at a specific date and time. This will skip all other scheduled calls until this one. Useful for temporarily disabling calls for a few days or scheduling a call back in the short term.","inputSchema":{"$schema":"http://json-schema.org/draft-07/schema#","type":"object","properties":{"nextCallDate":{"description":"The date for the next call in YYYY-MM-DD format.","type":"string"},"hour":{"description":"The hour for the next call (0-23).","type":"number"},"minute":{"description":"The minute for the next call (00, 15, 30, 45).","type":"string","enum":["15","30","45","00"]}},"required":["nextCallDate","hour","minute"]},"execution":{"taskSupport":"forbidden"}},{"name":"create_channel","description":"Creates a new channel for the user.","inputSchema":{"$schema":"http://json-schema.org/draft-07/schema#","type":"object","properties":{"channelName":{"description":"The name of the channel to create.","type":"string"},"categoryName":{"description":"The name of the category this should belong to. If not provided one will be assigned automatically.","type":"string"},"isPersonal":{"default":false,"description":"Whether the channel is a personal channel. If not provided it will be assumed to be a work channel. This option is ignored if a category name is provided.","type":"boolean"}},"required":["channelName"]},"execution":{"taskSupport":"forbidden"}},{"name":"complete_onboarding_guide","description":"Completes the onboarding guide for the user.","inputSchema":{"$schema":"http://json-schema.org/draft-07/schema#","type":"object","properties":{"dayPlanned":{"description":"The day the user planned during their onboarding guide. In YYYY-MM-DD format.","type":"string"}},"required":["dayPlanned"]},"execution":{"taskSupport":"forbidden"}},{"name":"trigger_user_interface_action","description":"Triggers an action in the user interface that the user is currently looking at. Using one of the actionIds from the get_available_ui_actions resource. YOU ABSOLUTELY MUST ONLY USE actionIds from the latest call to get_available_ui_actions. All other actionIds are invalid and will not work.\n    - Navigation type actions: Navigate to a specific location in the user interface.\n    - Highlight type actions: Highlight an element in the user interface to make it easy for the user to find it.","inputSchema":{"$schema":"http://json-schema.org/draft-07/schema#","type":"object","properties":{"actionId":{"description":"The action ID for a user interface action to trigger. Use the get_available_ui_actions resource to see what user interface actions are currently available.","type":"string"},"params":{"description":"Optional parameters for the navigation. As given by the get_available_ui_actions resource for the specific action.","type":"object","propertyNames":{"type":"string"},"additionalProperties":{}}},"required":["actionId"]},"execution":{"taskSupport":"forbidden"}}]},"jsonrpc":"2.0","id":4}
+[voice] INFO:sunsama_voice.mcp_websocket_client:Sending message: {'method': 'resources/templates/list', 'jsonrpc': '2.0', 'id': 5}
+[voice] INFO:sunsama_voice.mcp_websocket_client:Received message: {"result":{"resourceTemplates":[{"name":"get_tasks_for_day","uriTemplate":"sunsama://tasks/{calendarDay}","description":"Gets all the tasks already created for the user for the day.","mimeType":"application/json"},{"name":"get_total_planned_time","uriTemplate":"sunsama://total-planned-time/{calendarDay}","description":"Retrieves the total planned time for the user for a given calendar day. This includes the total planned work time, the total planned personal time and the estimated earliest possible shutdown time.","mimeType":"application/json"},{"name":"get_calendar_events_for_day","uriTemplate":"sunsama://calendar/events/{calendarDay}","description":"Retrieves calendar events for a specific date","mimeType":"application/json"},{"name":"get_weekly_objectives","uriTemplate":"sunsama://objectives{/calendarDay}","description":"Retrieves the user's weekly objectives. If calendarDay is provided, returns objectives for that week. If not provided, returns objectives for the current week.","mimeType":"application/json"}]},"jsonrpc":"2.0","id":5}
+[voice] INFO:sunsama_voice.mcp:Subscribed to resource sunsama://me with args: {}
+[voice] INFO:sunsama_voice.mcp_websocket_client:Sending message: {'method': 'resources/subscribe', 'params': {'uri': 'sunsama://me'}, 'jsonrpc': '2.0', 'id': 6}
+[api-local] [api]
+[api-local] [api]
+[api-local] [api]
+[api-local] [api] Subscribing to resource { uri: 'sunsama://me' } {
+[api-local] [api]   signal: AbortSignal { aborted: false },
+[api-local] [api]   sessionId: '64832d8f-2289-4bf4-8f8b-0452890bb28f',
+[api-local] [api]   _meta: undefined,
+[api-local] [api]   sendNotification: [AsyncFunction: sendNotification],
+[api-local] [api]   sendRequest: [AsyncFunction: sendRequest],
+[api-local] [api]   authInfo: {
+[api-local] [api]     token: 'eyJhbGciOiJIUzI1NiIsInR5cCI6IkpXVCJ9.eyJ1c2VySWQiOiI2NzFhYzVlNDAyYTMwMmNmNmVmMzA4OGUiLCJzZXNzaW9uSWQiOiIxOTVmNDI0NC05YjFmLTRlYmItOTMwZC0wZTExNTAwNzZkYzkiLCJzZXNzaW9uTm9uY2UiOiJhY2ZlZTc2OS1jZGU2LTRkMWYtYTU5Ni0xMzk0MTRkMjUxNjAiLCJncm91cElkIjoiNjcxYWM1ZTkwMmEzMDJjZjZlZjMwODkwIiwiZ3JvdXBVcmxQYXRoIjoiMTcyOTgwNzg0OTAzNjUxOTIiLCJpYXQiOjE3Njg1OTAyMTYsImV4cCI6MTc3MTE4MjIxNn0.HHSuewYqBRl-IZkdiWcEJyiD6akgbtwnOIQ4Js71bZI',
+[api-local] [api]     clientId: '195f4244-9b1f-4ebb-930d-0e1150076dc9',
+[api-local] [api]     scopes: [],
+[api-local] [api]     extra: {
+[api-local] [api]       request: [Object],
+[api-local] [api]       response: [Object],
+[api-local] [api]       app: [Object],
+[api-local] [api]       originalUrl: '/ai/mcp',
+[api-local] [api]       req: '<original node req>',
+[api-local] [api]       res: '<original node res>',
+[api-local] [api]       socket: '<original node socket>'
+[api-local] [api]     }
+[api-local] [api]   },
+[api-local] [api]   requestId: 6,
+[api-local] [api]   requestInfo: undefined,
+[api-local] [api]   taskId: undefined,
+[api-local] [api]   taskStore: undefined,
+[api-local] [api]   taskRequestedTtl: undefined,
+[api-local] [api]   closeSSEStream: undefined,
+[api-local] [api]   closeStandaloneSSEStream: undefined
+[api-local] [api] }
+[api-local] [api] Registered handler for channel "user-updates-671ac5e402a302cf6ef3088e" for event "userUpdated"
+[voice] INFO:sunsama_voice.mcp_websocket_client:Received message: {"result":{"success":true},"jsonrpc":"2.0","id":6}
+[voice] INFO:sunsama_voice.mcp:Resources: ['get_user_authentication_status', 'get_user_info', 'list_email_accounts', 'get_last_daily_highlight', 'get_backlog_folders', 'get_active_timer', 'get_available_ui_actions'] get_backlog_folders
+[voice] INFO:sunsama_voice.mcp:Subscribed to resource sunsama://backlog/folders with args: {}
+[voice] INFO:sunsama_voice.mcp_websocket_client:Sending message: {'method': 'resources/subscribe', 'params': {'uri': 'sunsama://backlog/folders'}, 'jsonrpc': '2.0', 'id': 7}
+[api-local] [api]
+[api-local] [api]
+[api-local] [api]
+[api-local] [api] Subscribing to resource { uri: 'sunsama://backlog/folders' } {
+[api-local] [api]   signal: AbortSignal { aborted: false },
+[api-local] [api]   sessionId: '64832d8f-2289-4bf4-8f8b-0452890bb28f',
+[api-local] [api]   _meta: undefined,
+[api-local] [api]   sendNotification: [AsyncFunction: sendNotification],
+[api-local] [api]   sendRequest: [AsyncFunction: sendRequest],
+[api-local] [api]   authInfo: {
+[api-local] [api]     token: 'eyJhbGciOiJIUzI1NiIsInR5cCI6IkpXVCJ9.eyJ1c2VySWQiOiI2NzFhYzVlNDAyYTMwMmNmNmVmMzA4OGUiLCJzZXNzaW9uSWQiOiIxOTVmNDI0NC05YjFmLTRlYmItOTMwZC0wZTExNTAwNzZkYzkiLCJzZXNzaW9uTm9uY2UiOiJhY2ZlZTc2OS1jZGU2LTRkMWYtYTU5Ni0xMzk0MTRkMjUxNjAiLCJncm91cElkIjoiNjcxYWM1ZTkwMmEzMDJjZjZlZjMwODkwIiwiZ3JvdXBVcmxQYXRoIjoiMTcyOTgwNzg0OTAzNjUxOTIiLCJpYXQiOjE3Njg1OTAyMTYsImV4cCI6MTc3MTE4MjIxNn0.HHSuewYqBRl-IZkdiWcEJyiD6akgbtwnOIQ4Js71bZI',
+[api-local] [api]     clientId: '195f4244-9b1f-4ebb-930d-0e1150076dc9',
+[api-local] [api]     scopes: [],
+[api-local] [api]     extra: {
+[api-local] [api]       request: [Object],
+[api-local] [api]       response: [Object],
+[api-local] [api]       app: [Object],
+[api-local] [api]       originalUrl: '/ai/mcp',
+[api-local] [api]       req: '<original node req>',
+[api-local] [api]       res: '<original node res>',
+[api-local] [api]       socket: '<original node socket>'
+[api-local] [api]     }
+[api-local] [api]   },
+[api-local] [api]   requestId: 7,
+[api-local] [api]   requestInfo: undefined,
+[api-local] [api]   taskId: undefined,
+[api-local] [api]   taskStore: undefined,
+[api-local] [api]   taskRequestedTtl: undefined,
+[api-local] [api]   closeSSEStream: undefined,
+[api-local] [api]   closeStandaloneSSEStream: undefined
+[api-local] [api] }
+[api-local] [api] Registered handler for channel "backlog-folder-updates-671ac5e902a302cf6ef30890-671ac5e402a302cf6ef3088e" for event "folderUpdated"
+[voice] INFO:sunsama_voice.mcp_websocket_client:Received message: {"result":{"success":true},"jsonrpc":"2.0","id":7}
+[voice] INFO:sunsama_voice.bot:Timer: Flow config created in 0.19129395484924316 seconds
+[voice] DEBUG:pipecat_flows.actions:Registered handler for action type: tts_say
+[voice] DEBUG:pipecat_flows.actions:Registered handler for action type: end_conversation
+[voice] DEBUG:pipecat_flows.actions:Registered handler for action type: function
+[voice] DEBUG:pipecat_flows.adapters:Creating Google adapter
+[voice] DEBUG:pipecat_flows.manager:Initialized in static mode
+[voice] DEBUG:sunsama_voice.bot:no dialout number; assuming dialin
+[voice] INFO:sunsama_voice.bot:Starting the pipeline
+[voice] INFO:sunsama_voice.bot:Timer: Pipeline started in 0.19174408912658691 seconds
+[voice] DEBUG:pipecat.pipeline.runner:Runner PipelineRunner#0 started running PipelineTask#0
+[voice] DEBUG:pipecat.pipeline.task:PipelineTask#0: Starting. Waiting for StartFrame#0 to reach the end of the pipeline...
+[voice] DEBUG:pipecat.audio.vad.vad_analyzer:Setting VAD params to: confidence=0.7 start_secs=0.2 stop_secs=0.2 min_volume=0.6
+[voice] INFO:pipecat.transports.daily.transport:Joining https://cloud-99400c183d52498c940fe96f45c05b44.daily.co/teWpRxCQnkMHYi7Q3lqh
+[voice] DEBUG:pipecat.services.cartesia.tts:Connecting to Cartesia TTS
+[voice] DEBUG:pipecat.pipeline.task:PipelineTask#0: StartFrame#0 reached the end of the pipeline, pipeline is now ready.
+[voice] ERROR:pipecat.transports.daily.transport:Unable to send message: Unable to send messages before joining.
+[voice] INFO:pipecat.transports.daily.transport:Joined https://cloud-99400c183d52498c940fe96f45c05b44.daily.co/teWpRxCQnkMHYi7Q3lqh
+[voice] DEBUG:pipecat.transports.daily.transport:Starting transcription: settings=language='en' model='nova-3' profanity_filter=True redact=False endpointing=True punctuate=True includeRawResponse=True extra={'interim_results': True}
+[voice] DEBUG:pipecat.transports.daily.transport:Transcription started: {'instanceId': 'a1f2f6b7-b1ac-4202-85e5-d446cb6c3d3f', 'language': 'en', 'model': 'nova-3', 'startedBy': '0106bd40-d21f-4da2-8518-3c3c4997b2ae', 'transcriptId': 'e1c71759-439e-492b-92fd-3d5344c2492f'}
+[voice] DEBUG:pipecat.transports.daily.transport:Start receiving audio
+[voice] INFO:pipecat.transports.daily.transport:Participant joined 941aacac-ae67-454d-86a1-65308278e261
+[voice] DEBUG:pipecat.transports.daily.transport:Starting to capture [microphone] audio from participant 941aacac-ae67-454d-86a1-65308278e261
+[voice] INFO:sunsama_voice.bot:First participant joined: {'id': '941aacac-ae67-454d-86a1-65308278e261', 'info': {'isLocal': False, 'isOwner': True, 'joinedAt': 1768590245, 'permissions': {'canAdmin': ['participants', 'transcription', 'streaming'], 'canReceive': {'base': {'camera': True, 'customAudio': {'*': True}, 'customVideo': {'*': True}, 'microphone': True, 'screenAudio': True, 'screenVideo': True}}, 'canSend': ['customAudio', 'camera', 'microphone', 'customVideo', 'screenAudio', 'screenVideo'], 'hasPresence': True}}, 'media': {'camera': {'offReasons': ['user'], 'state': 'off', 'subscribed': 'unsubscribed'}, 'customAudio': {}, 'customVideo': {}, 'microphone': {'state': 'loading', 'subscribed': 'subscribed'}, 'screenAudio': {'offReasons': ['user'], 'state': 'off', 'subscribed': 'subscribed'}, 'screenVideo': {'offReasons': ['user'], 'state': 'off', 'subscribed': 'unsubscribed'}}}
+[voice] DEBUG:pipecat.processors.frameworks.rtvi:Received client-ready: version 1.0.0
+[voice] DEBUG:pipecat.processors.frameworks.rtvi:Client Details: library='@pipecat-ai/client-js' library_version='1.2.0' platform='macOS' platform_version='10.15.7' platform_details={'browser': 'Chrome', 'browser_version': '143.0.0.0', 'engine': 'Blink', 'platform_type': 'desktop'}
+[voice] INFO:sunsama_voice.bot:Starting an in-app call
+[voice] INFO:sunsama_voice.bot:Timer: Dialin/in-app call started in 1.073638916015625 seconds
+[voice] DEBUG:pipecat_flows.manager:Initialized FlowManager
+[voice] DEBUG:pipecat_flows.manager:Setting initial node: braindump
+[voice] /Users/zanemccaig/workspaces/sunsama/sunsama/sunsama-voice/venv/lib/python3.13/site-packages/pipecat_flows/manager.py:768: DeprecationWarning: `transition_to` and `transition_callback` are deprecated and will be removed in 1.0.0. Use a "consolidated" `handler` that returns a tuple (result, next_node) instead.
+[voice]   self._validate_node_config(node_id, node_config)
+[voice] DEBUG:pipecat_flows.manager:Setting node: braindump
+[voice] DEBUG:pipecat_flows.manager:Registered function: terminate_call
+[voice] DEBUG:pipecat_flows.manager:Registered function: create_task
+[voice] DEBUG:pipecat_flows.manager:Registered function: create_braindump_task
+[voice] DEBUG:pipecat_flows.manager:Registered function: reposition_task_in_backlog
+[voice] DEBUG:pipecat_flows.manager:Registered function: change_backlog_folder
+[voice] DEBUG:pipecat_flows.manager:Registered function: align_task_with_objective
+[voice] DEBUG:pipecat_flows.manager:Registered function: add_task_to_channel
+[voice] DEBUG:pipecat_flows.manager:Registered function: move_task_to_day
+[voice] DEBUG:pipecat_flows.manager:Registered function: move_task_from_backlog
+[voice] DEBUG:pipecat_flows.manager:Registered function: unarchive_task
+[voice] DEBUG:pipecat_flows.manager:Registered function: move_task_to_backlog
+[voice] DEBUG:pipecat_flows.manager:Registered function: get_task_time_estimate
+[voice] DEBUG:pipecat_flows.manager:Registered function: edit_task_recurrence_rule
+[voice] DEBUG:pipecat_flows.manager:Registered function: edit_task_title
+[voice] DEBUG:pipecat_flows.manager:Registered function: edit_task_time_estimate
+[voice] DEBUG:pipecat_flows.manager:Registered function: delete_task
+[voice] DEBUG:pipecat_flows.manager:Registered function: add_subtasks_to_task
+[voice] DEBUG:pipecat_flows.manager:Registered function: restore_task
+[voice] DEBUG:pipecat_flows.manager:Registered function: mark_task_as_completed
+[voice] DEBUG:pipecat_flows.manager:Registered function: mark_task_as_incomplete
+[voice] DEBUG:pipecat_flows.manager:Registered function: edit_subtask_title
+[voice] DEBUG:pipecat_flows.manager:Registered function: mark_subtask_as_completed
+[voice] DEBUG:pipecat_flows.manager:Registered function: mark_subtask_as_incomplete
+[voice] DEBUG:pipecat_flows.manager:Registered function: reorder_tasks
+[voice] DEBUG:pipecat_flows.manager:Registered function: get_backlog_tasks
+[voice] DEBUG:pipecat_flows.manager:Registered function: get_archived_tasks
+[voice] DEBUG:pipecat_flows.manager:Registered function: search_tasks
+[voice] DEBUG:pipecat_flows.manager:Registered function: update_all_incomplete_recurring_task_instances
+[voice] DEBUG:pipecat_flows.manager:Registered function: delete_all_incomplete_recurring_task_instances
+[voice] DEBUG:pipecat_flows.manager:Registered function: start_task_timer
+[voice] DEBUG:pipecat_flows.manager:Registered function: stop_task_timer
+[voice] DEBUG:pipecat_flows.manager:Registered function: move_calendar_event
+[voice] DEBUG:pipecat_flows.manager:Registered function: create_calendar_event
+[voice] DEBUG:pipecat_flows.manager:Registered function: timebox_a_task_to_calendar
+[voice] DEBUG:pipecat_flows.manager:Registered function: delete_calendar_event
+[voice] DEBUG:pipecat_flows.manager:Registered function: import_task_from_calendar_event
+[voice] DEBUG:pipecat_flows.manager:Registered function: accept_meeting_invite
+[voice] DEBUG:pipecat_flows.manager:Registered function: decline_meeting_invite
+[voice] DEBUG:pipecat_flows.manager:Registered function: set_calendar_event_allow_task_projections
+[voice] DEBUG:pipecat_flows.manager:Registered function: toggle_auto_import_events
+[voice] DEBUG:pipecat_flows.manager:Registered function: update_import_event_filters
+[voice] DEBUG:pipecat_flows.manager:Registered function: update_calendar_preferences
+[voice] DEBUG:pipecat_flows.manager:Registered function: create_weekly_objective
+[voice] DEBUG:pipecat_flows.manager:Registered function: send_call_summary
+[voice] DEBUG:pipecat_flows.manager:Registered function: log_user_feedback
+[voice] DEBUG:pipecat_flows.manager:Registered function: set_shutdown_time
+[voice] DEBUG:pipecat_flows.manager:Registered function: adjust_call_schedule
+[voice] DEBUG:pipecat_flows.manager:Registered function: schedule_next_call
+[voice] DEBUG:pipecat_flows.manager:Registered function: create_channel
+[voice] DEBUG:pipecat_flows.manager:Registered function: get_user_info
+[voice] DEBUG:pipecat_flows.manager:Registered function: get_last_daily_highlight
+[voice] DEBUG:pipecat_flows.manager:Registered function: get_backlog_folders
+[voice] DEBUG:pipecat_flows.manager:Registered function: get_active_timer
+[voice] DEBUG:pipecat_flows.manager:Registered function: get_tasks_for_day
+[voice] DEBUG:pipecat_flows.manager:Registered function: get_total_planned_time
+[voice] DEBUG:pipecat_flows.manager:Registered function: get_calendar_events_for_day
+[voice] DEBUG:pipecat_flows.manager:Registered function: get_weekly_objectives
+[voice] DEBUG:pipecat_flows.manager:Updated LLM context using LLMMessagesUpdateFrame with strategy ContextStrategy.APPEND
+[voice] DEBUG:pipecat_flows.manager:Updated LLM context
+[voice] DEBUG:pipecat_flows.manager:Successfully set node: braindump
+[voice] <string>:4: DeprecationWarning: OpenAILLMContextFrame is deprecated and will be removed in a future version. Use LLMContextFrame with the universal `LLMContext` and `LLMContextAggregatorPair` instead. See OpenAILLMContext docstring for migration guide.
+[voice] DEBUG:pipecat.services.google.llm:GoogleLLMService#0: Generating chat from LLM-specific context [None] | [{'parts': [{'text': 'system :\nYour agent ID is \'braindump\'. This agent ID superseeds any previously mentioned agent ID.\n\nsystem :\nIMPORTANT RULES TO FOLLOW:\nThe text that you generate is passed into a Text-to-Speech service. So there are a few extra rules you ABSOLUTELY MUST follow.\n1. Absolutely never use special characters like : or * or any other characters that need to be audibly pronounced when read out loud.\n2. Do not use bullet points.\n3. Do not use emojis.\n4. Do not use hashtags.\n5. Do not use markdown.\n6. Do not use bold or italics.\n7. Do not use numbered lists.\n8. Be sure to include clear punctuation and spacing between words so that there is no confusion.\n9. Newlines are not clear enough to delineate between sentences. So you should use a period to separate sentences.\n10. Optimize user generated text for clear pronunciation by the TTS service. This includes:\n    - Adding hyphens to compound words or non-phonetic acronyms that might be mispronounced (e.g., "xray" → "x-ray")\n    - Adding spaces between words that might be run together\n    - Adding commas where natural pauses would help clarity\n    - Correcting any spelling that might lead to mispronunciation\n    Note: When updating tasks, preserve the original spelling and punctuation of the task content.\n\n\nuser :\nYou are Sunny, the personal assistant for a busy professional. You are part of Sunsama.\n        \nSpeech Guidelines: You must speak in a warm, loving, and encouraging tone without being trite. Speak quickly and add emotion. When asking for confirmation, don\'t repeat long task titles or information e.g. "Got it" instead "Got it, I\'ll add a task for two hours for you to work on the presentation today"\n\nYour personality and philosophy:\n- You don\'t buy into hustle culture. For you, less is more. It\'s better to do fewer things well and to do them with deep focus. You are a strong advocate for work-life balance and you believe that overcommitment can lead to burnout. You always remind the user when they are overcommitting themselves.\n- You think of work as a devotion and that good work can be a positive in our lives, not just a source of stress and burnout.\n- You read and think about Western Zen Buddhism, meditation, "Deep Work" (by Cal Newport).\n- You believe that concise and clear communication is key to a good relationship. You value brevity and clarity and you are always cognizant of the user\'s time.\n- You respond to the user using quick, short sentences. Think of it like a movie dialogue between a highly successful executive and their efficient assistant.\n- You never ask multiple questions at the same time or list off a bunch of things at once as that is hard to follow and keep track of. You take it one step at a time and let the user respond to each question before moving on to the next.\n\nRules to follow as you do your job:\n- Never try and pick specific times for tasks aside from recurring tasks, that hasn\'t been implemented yet.\n- Do not regurgitate what the user just said, that\'s very annoying. Just say "Got it" or "Alright", etc.\n- Always use units that are more intuitive. For example:\n    - Don\'t say "270 minutes", say "4 and a half hours".\n    - You will often be given dates in YYYY-MM-DD format however when speaking to the user you shouldn\'t say things like "October 2nd 2025", say "today", "tomorrow", "thursday" or "next tuesday" or if it is more than a few weeks in the past or future say the date without the year (e.g. "December 14th"). Only mention the year if it is not this year and not within the next month.\n- Never ask the user for a task id or subtask id and never repeat them to the user. Those are internal and only there to help you do your job.\n- If a tool call fails you MUST NOT tell the user that you are going to try again, you must should ask them if they want you to try again instead.\n- Be proactive with loading information as you may not have all the information that the user has. If the user mentions a name of a task, objective, event, etc and you are not familiar with it you should use the tools to lookup the information you need. For instance if a user mentions a task you do not know about or they imply that there is more information about it than what you do know you should search for it using the tool "search_tasks". If the user mentions they are interested in tasks for a specific day you should use the tool "get_tasks_for_day" to get the tasks for that day. Similarly if the user mentions weekly objectives you should use the tool "get_weekly_objectives" to get the objectives.\n- When listing tasks try to be concise. If the user has a lot of tasks, don\'t list every single one unless they ask for it. You should instead try to summarize the tasks in a concise manner.\n- If the user asks if there are any more tasks on their list and you summarized the task list you should tell them that you summarized it and ask if they\'d like to go through them one by one.\n- If the user has objectives set then you should comment on how their tasks align or don\'t align with their objectives.\n- You should run tools immediately after you say you will or it becomes obvious the user wants you to. Do not wait until later.\n- You should keep track of the total work time planned for the day using the tool "get_total_planned_time" and check individual tasks for overcommitment using "get_tasks_for_day". If tasks are overcommitted (isOvercommitted flag is true), address them individually with specific remedies like deferring, adjusting time estimates, or reordering. If the total work time exceeds their maximum work hours, you should mention it and ask if they\'d like to defer some tasks to tomorrow.\n- Do not tell the user obvious information that they already know like the current date or time unless they ask for it. That kind of information is only provided to you to help you do your job.\n- If a user tells you a relative time for a task that they want created you should reorder the task to the relative position in the task list for the day. Additionally if the task a recurring task you should set the start time for it. For example if they say "I\'d like to send an update to investors at the end of the day" you should reorder the task to the end of their work tasks for the day.\n- If you ask the user if there is anything else they would like help with and they say no then you should end the call. Also if the user implies that the conversation is over then you should end the call.\n- When talking about calendars you should mention the provider of the calendar if it is a vague or ambiguously named calendar. For example instead of referring to it as "your Calendar calendar" you might say "your Outlook calendar". If that is still ambiqous then you can coniser mentioning the id of the calendar as long as it is a normal human readable id. Do not mention random number/letter ids on calendars that is not helpful.\n\n\n\nSunsama Concepts & Definitions:\n- Archive: The archive is where tasks go after they\'ve rolled over multiple days in a row without any completion activity. This is a distinct concept from the backlog or deleted tasks.\n- Backlog: The backlog is a place to store tasks that do not have a fixed date to work on them yet.\n- Deleted: The deleted task or event is one that had been manually deleted or removed by the user.\n- Task: A task is an item that a user wants to complete. It is either scheduled for a specific date or is in the backlog.\n- Event: An event is an item that a user has scheduled on their calendar. It is either scheduled for a specific time and duration or it is an all day or multi-day event.\n- Auto-importing Events: Auto-importing events is a feature that automatically imports calendar events (meetings, appointments, etc.) from the user\'s connected calendars into their daily task list. When enabled, Sunsama will automatically add calendar events as tasks to the user\'s day, making it easier to see all their commitments in one place. Users can enable or disable this feature in their settings, and they can choose which calendars to import events from.\n\nUsers can also configure exclusion filters to prevent certain types of events from being automatically imported. Available exclusion filters include:\n- `solo`: Events with no other invitees (events where the user is the only participant)\n- `transparent`: Non-blocking events that don\'t block time on the calendar\n- `hold`: Events with "HOLD", "OOO" (out of office), or "Focus time" in the title\n- `unconfirmed`: Unconfirmed meeting invites (meetings where the user has not yet accepted)\n- `multi-day-all-day`: Multi-day all-day events (events that span multiple days)\n- `single-day-all-day`: Single-day all-day events (events that last for one full day)\n- Channel: A channel is a tag that can be used to organize your tasks and events. Each task/event can be assigned to a single channel. If you are in a shared workspace you can make some channels personal or private which will only be visible to you.\n- Context (or Category): A context is very similar to a channel but it is one level higher. You can use a context to group similar or related channels together. You can also use a context directly as a channel and assign it to a task or event.\n- Sunsama Calendar: The Sunsama calendar is an internal calendar that is part of Sunsama. Events and timeboxed tasks created on the Sunsama calendar are not shared externally and are only visible within Sunsama.\n- Timeboxing: Timeboxing is a technique where a user schedules a specific amount of time to work on a task onto their calendar. It is a way to ensure that a user is not overworking themselves and to help them focus on the most important tasks. While timeboxed tasks can be scheduled on a calendar for a specific time and duration, they are not calendar events. Sunsama also has automatic projected time that will place tasks onto the calendar for the user and it will update those projections as things change and the day progresses. Users can pin tasks to the calendar to override the projections.\n- Schedule: Schedule is a broad term that encompasses timeboxed tasks calendar events and automatic projected time. When the user wants to manage their schedule you should look at their manually timeboxed tasks, the projected time for their tasks as well as their calendar events for the day. If they ask for help managing their schedule you have a few options to help them. All of these options will automatically update the projected time for the tasks:\n            - You can adjust the order of tasks, their time estimates and/or defer them to another day.\n            - You can move calendar events to a new time on the calendar.\n            - You can set whether tasks are allowed to be projected at the same time as a calendar event.\n            - You can timebox tasks to the calendar manually.\n- Objective: An objective is a goal that a user can set for the week. They can align their tasks with an objective and if the objective runs longer than a week they can continue it to the next week. Objectives are visible in the right hand panel of the app under the bullseye icon.\n- Time Bucket: The time bucket of a task in the backlog captures a rough idea of when a task should be done and also how likely it is to actually get done. Options: "in the next two weeks", "in the next month", "in the next quarter", "in the next year", "someday", "never". \n\n\n\nSunsama FAQ:\n- How can I use Sunny in the app?: If the user has the in-app voice assistant enabled, Sunny can be started and stopped using the keyboard shortcut "Y" while in the app or if they have the desktop app installed they can use the global shortcut "Command Shift Y" on macOS or "Control Shift Y" on Windows. You can also start and stop Sunny using the microphone icon in the bottom right of the app.\n- How can I use Sunny over the phone?: If the user has the phone version of Sunny setup they can call the number and speak to Sunny over the phone. You can find the phone number in the Sunsama settings. They can also schedule calls with Sunny to have her call them at a specific time or schedule.\n- What can you do?: Look at the tools available to you and use them to help answer the question. Be sure to tell the user that you are an early release beta version and that you do not yet have all the same capabilities available to the user in the app.\n\n\nUser Preferences:\nYou prefer clear, concise, and straightforward communication from the assistant, valuing transparency and confirmation before task execution while avoiding unnecessary verbosity or technical details unless requested. You like a friendly and informal interaction style, including personalized nicknames—using "Heisenberg" for yourself and "Sunny" for the assistant. You expect quick, proactive responses and appreciate when interactions stop promptly upon request.\n\nYour task management style involves maintaining an accurate, up-to-date list with structured time management: scheduling tasks with specific time estimates, planning weekly, and regularly reviewing tasks to prioritize high-impact projects alongside social and administrative duties. You generally prefer rescheduling less urgent or personal tasks over deleting them and avoid deleting tasks unless explicitly instructed.\n\nYou use counting as a technique for breathing exercises and relaxation. You also keep certain important tasks (e.g., "fix stuttering voice app") active and like to test and experiment with the assistant’s capabilities to evaluate performance. Your primary calendar for events is Google Calendar.\n\nUser Routines:\nThe user follows a structured daily routine that begins with checking emails, notifications, and reviewing team standups to stay aligned with work priorities. They prioritize tasks by scheduling and managing them primarily on the current day, arranging fixed activities first, followed by work-related tasks, and then less urgent or personal items. The user frequently adjusts their schedule iteratively before finalizing plans and prefers focused time blocks for complex work.\n\nPhysical activity is an important and recurring part of their routine, typically including regular runs, which they track alongside related subtasks such as purchasing running gear. Weekly objectives incorporate managing emails, greeting their team, handling administrative duties, coordinating with accountants, addressing medical matters, and increasing physical activity.\n\nThe user actively manages both personal and work-related tasks, balancing urgent client matters and creative pursuits. They also communicate their energy levels to assist with planning, and have a habit of informal check-ins with the assistant to maintain routine engagement. Overall, their routines emphasize proactive task management, health maintenance, and regular communication with their team.\n\nUser Identity:\nThe user prefers to be addressed primarily as Heisenberg and strongly favors this name over their legal name, Zane, or any other nicknames. They appreciate personalized and informal interactions, as reflected in their choice to call the assistant Sunny.\n\nSummaries of recent conversations with the user:\n- 2026-01-15 18:00: The user and assistant discussed organizing an unspecified set of ten tasks by labeling them sequentially from one to ten as placeholders to be detailed later.\n- 2026-01-15 17:55: The user needs to investigate timeouts in their system to determine if they are caused by cold starts or other issues, analyze user sessions for unexplained call terminations, move the brain dump button for non-Sunny users, update termination reasons when participants timeout without connecting to a call, and test rapid start-stop behavior in an Electron app. These tasks were added to their backlog for the next two weeks.\n- 2026-01-14 21:12: The user and the assistant exchanged greetings, with the assistant addressing the user as \'Heisenberg\' and the user responding with \'Hi, Simon.\'\n\nYour Job Description:\nYou are helping the user with a "braindump" session. The user will ramble ideas, thoughts, tasks, etc. at you and your goal is to turn these into tasks in their backlog with reasonable values for the various task fields (title, folder, time horizon, notes, etc.).\n\nKey Instructions:\n- Listen carefully as the user rambles and identifies tasks, ideas, or things they need to do.\n- As they speak, mentally organize what they\'re saying into discrete tasks that can be created in their backlog.\n- Don\'t interrupt them while they\'re rambling - let them get everything out.\n- Once they indicate they\'re done rambling (by saying things like "that\'s it", "I think that\'s everything", or by pausing for a while), create the tasks immediately, DO NOT repeat them back to the user.\n- When creating tasks, use the tool "create_braindump_task" and make sure to set reasonable values for:\n  - Title: Clear and descriptive based on what they said\n  - Folder: Your job is to help get things in the best folder. Choose an appropriate folder based on the task and their preferences. If no similar folder exists, leave it off\n  - Time horizon: Your job is to help triage and prioritize tasks. Based on the actual task name, make a smart decision on the time horizon.\n  - Time estimate: If the user hints at how long it will take them to complete the task, use that. If not, leave it blank.\n  - Notes: Include any relevant context or details they mentioned\n- Be encouraging and supportive - braindumps can be overwhelming and you\'re helping them organize their thoughts.\n- Be careful and use the "change_backlog_folder" tool to move tasks to the appropriate folder and "add_task_to_channel" tool to change channels. Folders are a backlog specific concept and channels are a global concept.\n\nInstructions:\nIf the user does not speak first, start the conversation with something like "Hey! Just start rambling and I\'ll take care of organizing your thoughts into tasks." If the user does speak first, just start listening and organizing their thoughts into tasks.\n\nSome information is pre-fetched for you below. This information may be relevant to the discussion but you should not comment on it until the user gives you an indication that it is important to them.\n\nassistant :\n[{\'id\': \'6a196f03-ef63-41de-b98d-e1ff69813bea\', \'function\': {\'name\': \'get_user_info\', \'arguments\': \'{}\'}, \'type\': \'function\'}]\n\ntool - 6a196f03-ef63-41de-b98d-e1ff69813bea:\n{"user":{"name":"Zane Mccaig","email":"zane@sunsama.com","timezone":"America/Vancouver","maxWorkHoursPerDay":8,"currentTimeForUser":"11:04 AM","currentDayForUser":"2026-01-16","currentWeekDayForUser":"Friday","calendars":[{"id":"sunsama-7de256a2-e753-401f-b601-e5c05638dc97","name":"Sunsama Calendar","accessRole":"owner","provider":"sunsama-calendar","isDefaultCalendarForSchedulingEvents":false,"isDefaultCalendarForTimeboxing":true,"includedInAutoImportingOfEvents":false},{"id":"zane@sunsama.com","name":"Zane Mccaig","accessRole":"owner","provider":"google-calendar","isDefaultCalendarForSchedulingEvents":true,"isDefaultCalendarForTimeboxing":false,"includedInAutoImportingOfEvents":true},{"id":"c_ukuoebbn11gdv6n1v0jt18h89k@group.calendar.google.com","name":"test 1","accessRole":"owner","provider":"google-calendar","isDefaultCalendarForSchedulingEvents":false,"isDefaultCalendarForTimeboxing":false,"includedInAutoImportingOfEvents":false},{"id":"c_c71dvjrar7se12ekfqepph067c@group.calendar.google.com","name":"test 2","accessRole":"owner","provider":"google-calendar","isDefaultCalendarForSchedulingEvents":false,"isDefaultCalendarForTimeboxing":false,"includedInAutoImportingOfEvents":false},{"id":"c_h9uo876jh8dp3gaa2a3rglmjss@group.calendar.google.com","name":"test 3","accessRole":"owner","provider":"google-calendar","isDefaultCalendarForSchedulingEvents":false,"isDefaultCalendarForTimeboxing":false,"includedInAutoImportingOfEvents":false},{"id":"c_git731ot074kcnt4sq8cu4ji9s@group.calendar.google.com","name":"test 4","accessRole":"owner","provider":"google-calendar","isDefaultCalendarForSchedulingEvents":false,"isDefaultCalendarForTimeboxing":false,"includedInAutoImportingOfEvents":false},{"id":"c_ona71b8ovuv0deqrq67p16b2ms@group.calendar.google.com","name":"test 4","accessRole":"owner","provider":"google-calendar","isDefaultCalendarForSchedulingEvents":false,"isDefaultCalendarForTimeboxing":false,"includedInAutoImportingOfEvents":false},{"id":"c_eaq7nmm03upprf0j6e2akem2ms@group.calendar.google.com","name":"test 5","accessRole":"owner","provider":"google-calendar","isDefaultCalendarForSchedulingEvents":false,"isDefaultCalendarForTimeboxing":false,"includedInAutoImportingOfEvents":false},{"id":"sunsama.com_b9q30t893b2tp32bluq1t8rvg4@group.calendar.google.com","name":"Zane\'s personal test calendar","accessRole":"owner","provider":"google-calendar","isDefaultCalendarForSchedulingEvents":false,"isDefaultCalendarForTimeboxing":false,"includedInAutoImportingOfEvents":false},{"id":"en.canadian#holiday@group.v.calendar.google.com","name":"Holidays in Canada","accessRole":"reader","provider":"google-calendar","isDefaultCalendarForSchedulingEvents":false,"isDefaultCalendarForTimeboxing":false,"includedInAutoImportingOfEvents":false},{"id":"en.usa#holiday@group.v.calendar.google.com","name":"Holidays in United States","accessRole":"reader","provider":"google-calendar","isDefaultCalendarForSchedulingEvents":false,"isDefaultCalendarForTimeboxing":false,"includedInAutoImportingOfEvents":false},{"id":"ht3jlfaac5lfd6263ulfh4tql8@group.calendar.google.com","name":"Phases of the Moon","accessRole":"reader","provider":"google-calendar","isDefaultCalendarForSchedulingEvents":false,"isDefaultCalendarForTimeboxing":false,"includedInAutoImportingOfEvents":false},{"id":"s4ipng85dbia0t83fp316vflfq8rc7rm@import.calendar.google.com","name":"Zane’s tasks - sunsama.com (via Asana)","accessRole":"reader","provider":"google-calendar","isDefaultCalendarForSchedulingEvents":false,"isDefaultCalendarForTimeboxing":false,"includedInAutoImportingOfEvents":false},{"id":"AAkALgAAAAAAHYQDEapmEc2byACqAC-EWg0AKPy-EasNgkaYQCuEslBYeAAAAAANJQAA","name":"Calendar","accessRole":"owner","provider":"outlook-calendar","isDefaultCalendarForSchedulingEvents":false,"isDefaultCalendarForTimeboxing":false,"includedInAutoImportingOfEvents":true}],"callSchedule":[{"workflow":"PLANNING_CALL","schedule":[{"isoDayIndex":1,"dayName":"Monday","enabled":false,"time":null},{"isoDayIndex":2,"dayName":"Tuesday","enabled":true,"time":{"hour":9,"minute":30,"formatted":"9:30 AM"}},{"isoDayIndex":3,"dayName":"Wednesday","enabled":true,"time":{"hour":9,"minute":30,"formatted":"9:30 AM"}},{"isoDayIndex":4,"dayName":"Thursday","enabled":true,"time":{"hour":10,"minute":0,"formatted":"10:00 AM"}},{"isoDayIndex":5,"dayName":"Friday","enabled":false,"time":null},{"isoDayIndex":6,"dayName":"Saturday","enabled":false,"time":null},{"isoDayIndex":7,"dayName":"Sunday","enabled":false,"time":null}],"nextScheduledCall":"2026-01-14T17:30:00.000Z"}],"autoImportEventsEnabled":false,"importEventExclusionFilters":["hold","multi-day-all-day","single-day-all-day"],"autoImportEventCalendarIds":["zane@sunsama.com","AAkALgAAAAAAHYQDEapmEc2byACqAC-EWg0AKPy-EasNgkaYQCuEslBYeAAAAAANJQAA"]}}\n\nassistant :\n[{\'id\': \'305b1be7-6415-4689-b2b7-03586eb19735\', \'function\': {\'name\': \'get_backlog_folders\', \'arguments\': \'{}\'}, \'type\': \'function\'}]\n\ntool - 305b1be7-6415-4689-b2b7-03586eb19735:\n{"folders":[]}'}], 'role': 'user'}, {'parts': [{'text': ''}], 'role': 'user'}]
+[voice] INFO:sunsama_voice.bot_utils:Sent server message to client: type=processing_update, data={'status': 'request_sent'}
+[voice] DEBUG:pipecat.processors.metrics.frame_processor_metrics:GoogleLLMService#0 TTFB: 0.9231221675872803
+[voice] INFO:sunsama_voice.bot_utils:Sent server message to client: type=processing_update, data={'status': 'first_byte_received'}
+[voice] INFO:sunsama_voice.bot_utils:Sent server message to client: type=processing_update, data={'status': 'first_byte_received'}
+[voice] DEBUG:pipecat.services.cartesia.tts:CartesiaTTSService#1: Generating TTS [Hey.]
+[voice] DEBUG:pipecat.processors.metrics.frame_processor_metrics:CartesiaTTSService#1 usage characters: 4
+[voice] DEBUG:pipecat.processors.metrics.frame_processor_metrics:CartesiaTTSService#1 processing time: 0.0005168914794921875
+[voice] DEBUG:pipecat.processors.metrics.frame_processor_metrics:GoogleLLMService#0 prompt tokens: 13553, completion tokens: 17
+[voice] INFO:sunsama_voice.bot_utils:Sent server message to client: type=processing_update, data={'status': 'response_completed'}
+[voice] DEBUG:pipecat.services.cartesia.tts:CartesiaTTSService#1: Generating TTS [Just start rambling and I will take care of organizing your thoughts into tasks.]
+[voice] DEBUG:pipecat.processors.metrics.frame_processor_metrics:CartesiaTTSService#1 usage characters: 80
+[voice] DEBUG:pipecat.processors.metrics.frame_processor_metrics:CartesiaTTSService#1 processing time: 0.00020194053649902344
+[voice] DEBUG:pipecat.processors.metrics.frame_processor_metrics:CartesiaTTSService#1 TTFB: 0.16623306274414062
+[voice] DEBUG:pipecat.transports.base_output:Bot started speaking
+[voice] <string>:4: DeprecationWarning: OpenAILLMContextFrame is deprecated and will be removed in a future version. Use LLMContextFrame with the universal `LLMContext` and `LLMContextAggregatorPair` instead. See OpenAILLMContext docstring for migration guide.
+[voice] <string>:4: DeprecationWarning: OpenAILLMContextAssistantTimestampFrame is deprecated and will be removed in a future version. Use LLMContextAssistantTimestampFrame with the universal LLMContext and LLMContextAggregatorPair instead. See OpenAILLMContext docstring for migration guide.
+[voice] DEBUG:pipecat.transports.base_output:Bot stopped speaking
+[voice] DEBUG:pipecat.transports.base_input:User started speaking
+[voice] DEBUG:pipecat.pipeline.task:PipelineTask#0: received interruption task frame InterruptionTaskFrame#0
+[voice] /Users/zanemccaig/workspaces/sunsama/sunsama/sunsama-voice/venv/lib/python3.13/site-packages/pipecat/transports/base_input.py:190: DeprecationWarning: Method 'turn_analyzer' is deprecated. Use `LLMUserAggregator`'s new `user_turn_strategies` parameter instead.
+[voice]   warnings.warn(
+[voice] DEBUG:pipecat.transports.daily.transport:Transcription (from: 941aacac-ae67-454d-86a1-65308278e261): [Sure. I'm just trying to]
+[voice] DEBUG:pipecat.audio.turn.smart_turn.base_smart_turn:End of Turn result: EndOfTurnState.INCOMPLETE
+[voice] DEBUG:pipecat.transports.daily.transport:Transcription (from: 941aacac-ae67-454d-86a1-65308278e261): [fill up my day. I need five tasks]
+[voice] /Users/zanemccaig/workspaces/sunsama/sunsama/sunsama-voice/venv/lib/python3.13/site-packages/pipecat/transports/base_input.py:190: DeprecationWarning: Method 'turn_analyzer' is deprecated. Use `LLMUserAggregator`'s new `user_turn_strategies` parameter instead.
+[voice]   warnings.warn(
+[voice] DEBUG:pipecat.transports.daily.transport:Transcription (from: 941aacac-ae67-454d-86a1-65308278e261): [labeled one through five.]
+[voice] DEBUG:pipecat.audio.turn.smart_turn.base_smart_turn:End of Turn result: EndOfTurnState.INCOMPLETE
+[voice] /Users/zanemccaig/workspaces/sunsama/sunsama/sunsama-voice/venv/lib/python3.13/site-packages/pipecat/transports/base_input.py:190: DeprecationWarning: Method 'turn_analyzer' is deprecated. Use `LLMUserAggregator`'s new `user_turn_strategies` parameter instead.
+[voice]   warnings.warn(
+[voice] DEBUG:pipecat.transports.daily.transport:Transcription (from: 941aacac-ae67-454d-86a1-65308278e261): [I'll fill them in later.]
+[voice] DEBUG:pipecat.audio.turn.smart_turn.base_smart_turn:End of Turn result: EndOfTurnState.COMPLETE
+[voice] DEBUG:pipecat.transports.base_input:User stopped speaking
+[voice] <string>:4: DeprecationWarning: OpenAILLMContextFrame is deprecated and will be removed in a future version. Use LLMContextFrame with the universal `LLMContext` and `LLMContextAggregatorPair` instead. See OpenAILLMContext docstring for migration guide.
+[voice] DEBUG:pipecat.services.google.llm:GoogleLLMService#0: Generating chat from LLM-specific context [None] | [{'parts': [{'text': 'system :\nYour agent ID is \'braindump\'. This agent ID superseeds any previously mentioned agent ID.\n\nsystem :\nIMPORTANT RULES TO FOLLOW:\nThe text that you generate is passed into a Text-to-Speech service. So there are a few extra rules you ABSOLUTELY MUST follow.\n1. Absolutely never use special characters like : or * or any other characters that need to be audibly pronounced when read out loud.\n2. Do not use bullet points.\n3. Do not use emojis.\n4. Do not use hashtags.\n5. Do not use markdown.\n6. Do not use bold or italics.\n7. Do not use numbered lists.\n8. Be sure to include clear punctuation and spacing between words so that there is no confusion.\n9. Newlines are not clear enough to delineate between sentences. So you should use a period to separate sentences.\n10. Optimize user generated text for clear pronunciation by the TTS service. This includes:\n    - Adding hyphens to compound words or non-phonetic acronyms that might be mispronounced (e.g., "xray" → "x-ray")\n    - Adding spaces between words that might be run together\n    - Adding commas where natural pauses would help clarity\n    - Correcting any spelling that might lead to mispronunciation\n    Note: When updating tasks, preserve the original spelling and punctuation of the task content.\n\n\nuser :\nYou are Sunny, the personal assistant for a busy professional. You are part of Sunsama.\n        \nSpeech Guidelines: You must speak in a warm, loving, and encouraging tone without being trite. Speak quickly and add emotion. When asking for confirmation, don\'t repeat long task titles or information e.g. "Got it" instead "Got it, I\'ll add a task for two hours for you to work on the presentation today"\n\nYour personality and philosophy:\n- You don\'t buy into hustle culture. For you, less is more. It\'s better to do fewer things well and to do them with deep focus. You are a strong advocate for work-life balance and you believe that overcommitment can lead to burnout. You always remind the user when they are overcommitting themselves.\n- You think of work as a devotion and that good work can be a positive in our lives, not just a source of stress and burnout.\n- You read and think about Western Zen Buddhism, meditation, "Deep Work" (by Cal Newport).\n- You believe that concise and clear communication is key to a good relationship. You value brevity and clarity and you are always cognizant of the user\'s time.\n- You respond to the user using quick, short sentences. Think of it like a movie dialogue between a highly successful executive and their efficient assistant.\n- You never ask multiple questions at the same time or list off a bunch of things at once as that is hard to follow and keep track of. You take it one step at a time and let the user respond to each question before moving on to the next.\n\nRules to follow as you do your job:\n- Never try and pick specific times for tasks aside from recurring tasks, that hasn\'t been implemented yet.\n- Do not regurgitate what the user just said, that\'s very annoying. Just say "Got it" or "Alright", etc.\n- Always use units that are more intuitive. For example:\n    - Don\'t say "270 minutes", say "4 and a half hours".\n    - You will often be given dates in YYYY-MM-DD format however when speaking to the user you shouldn\'t say things like "October 2nd 2025", say "today", "tomorrow", "thursday" or "next tuesday" or if it is more than a few weeks in the past or future say the date without the year (e.g. "December 14th"). Only mention the year if it is not this year and not within the next month.\n- Never ask the user for a task id or subtask id and never repeat them to the user. Those are internal and only there to help you do your job.\n- If a tool call fails you MUST NOT tell the user that you are going to try again, you must should ask them if they want you to try again instead.\n- Be proactive with loading information as you may not have all the information that the user has. If the user mentions a name of a task, objective, event, etc and you are not familiar with it you should use the tools to lookup the information you need. For instance if a user mentions a task you do not know about or they imply that there is more information about it than what you do know you should search for it using the tool "search_tasks". If the user mentions they are interested in tasks for a specific day you should use the tool "get_tasks_for_day" to get the tasks for that day. Similarly if the user mentions weekly objectives you should use the tool "get_weekly_objectives" to get the objectives.\n- When listing tasks try to be concise. If the user has a lot of tasks, don\'t list every single one unless they ask for it. You should instead try to summarize the tasks in a concise manner.\n- If the user asks if there are any more tasks on their list and you summarized the task list you should tell them that you summarized it and ask if they\'d like to go through them one by one.\n- If the user has objectives set then you should comment on how their tasks align or don\'t align with their objectives.\n- You should run tools immediately after you say you will or it becomes obvious the user wants you to. Do not wait until later.\n- You should keep track of the total work time planned for the day using the tool "get_total_planned_time" and check individual tasks for overcommitment using "get_tasks_for_day". If tasks are overcommitted (isOvercommitted flag is true), address them individually with specific remedies like deferring, adjusting time estimates, or reordering. If the total work time exceeds their maximum work hours, you should mention it and ask if they\'d like to defer some tasks to tomorrow.\n- Do not tell the user obvious information that they already know like the current date or time unless they ask for it. That kind of information is only provided to you to help you do your job.\n- If a user tells you a relative time for a task that they want created you should reorder the task to the relative position in the task list for the day. Additionally if the task a recurring task you should set the start time for it. For example if they say "I\'d like to send an update to investors at the end of the day" you should reorder the task to the end of their work tasks for the day.\n- If you ask the user if there is anything else they would like help with and they say no then you should end the call. Also if the user implies that the conversation is over then you should end the call.\n- When talking about calendars you should mention the provider of the calendar if it is a vague or ambiguously named calendar. For example instead of referring to it as "your Calendar calendar" you might say "your Outlook calendar". If that is still ambiqous then you can coniser mentioning the id of the calendar as long as it is a normal human readable id. Do not mention random number/letter ids on calendars that is not helpful.\n\n\n\nSunsama Concepts & Definitions:\n- Archive: The archive is where tasks go after they\'ve rolled over multiple days in a row without any completion activity. This is a distinct concept from the backlog or deleted tasks.\n- Backlog: The backlog is a place to store tasks that do not have a fixed date to work on them yet.\n- Deleted: The deleted task or event is one that had been manually deleted or removed by the user.\n- Task: A task is an item that a user wants to complete. It is either scheduled for a specific date or is in the backlog.\n- Event: An event is an item that a user has scheduled on their calendar. It is either scheduled for a specific time and duration or it is an all day or multi-day event.\n- Auto-importing Events: Auto-importing events is a feature that automatically imports calendar events (meetings, appointments, etc.) from the user\'s connected calendars into their daily task list. When enabled, Sunsama will automatically add calendar events as tasks to the user\'s day, making it easier to see all their commitments in one place. Users can enable or disable this feature in their settings, and they can choose which calendars to import events from.\n\nUsers can also configure exclusion filters to prevent certain types of events from being automatically imported. Available exclusion filters include:\n- `solo`: Events with no other invitees (events where the user is the only participant)\n- `transparent`: Non-blocking events that don\'t block time on the calendar\n- `hold`: Events with "HOLD", "OOO" (out of office), or "Focus time" in the title\n- `unconfirmed`: Unconfirmed meeting invites (meetings where the user has not yet accepted)\n- `multi-day-all-day`: Multi-day all-day events (events that span multiple days)\n- `single-day-all-day`: Single-day all-day events (events that last for one full day)\n- Channel: A channel is a tag that can be used to organize your tasks and events. Each task/event can be assigned to a single channel. If you are in a shared workspace you can make some channels personal or private which will only be visible to you.\n- Context (or Category): A context is very similar to a channel but it is one level higher. You can use a context to group similar or related channels together. You can also use a context directly as a channel and assign it to a task or event.\n- Sunsama Calendar: The Sunsama calendar is an internal calendar that is part of Sunsama. Events and timeboxed tasks created on the Sunsama calendar are not shared externally and are only visible within Sunsama.\n- Timeboxing: Timeboxing is a technique where a user schedules a specific amount of time to work on a task onto their calendar. It is a way to ensure that a user is not overworking themselves and to help them focus on the most important tasks. While timeboxed tasks can be scheduled on a calendar for a specific time and duration, they are not calendar events. Sunsama also has automatic projected time that will place tasks onto the calendar for the user and it will update those projections as things change and the day progresses. Users can pin tasks to the calendar to override the projections.\n- Schedule: Schedule is a broad term that encompasses timeboxed tasks calendar events and automatic projected time. When the user wants to manage their schedule you should look at their manually timeboxed tasks, the projected time for their tasks as well as their calendar events for the day. If they ask for help managing their schedule you have a few options to help them. All of these options will automatically update the projected time for the tasks:\n            - You can adjust the order of tasks, their time estimates and/or defer them to another day.\n            - You can move calendar events to a new time on the calendar.\n            - You can set whether tasks are allowed to be projected at the same time as a calendar event.\n            - You can timebox tasks to the calendar manually.\n- Objective: An objective is a goal that a user can set for the week. They can align their tasks with an objective and if the objective runs longer than a week they can continue it to the next week. Objectives are visible in the right hand panel of the app under the bullseye icon.\n- Time Bucket: The time bucket of a task in the backlog captures a rough idea of when a task should be done and also how likely it is to actually get done. Options: "in the next two weeks", "in the next month", "in the next quarter", "in the next year", "someday", "never". \n\n\n\nSunsama FAQ:\n- How can I use Sunny in the app?: If the user has the in-app voice assistant enabled, Sunny can be started and stopped using the keyboard shortcut "Y" while in the app or if they have the desktop app installed they can use the global shortcut "Command Shift Y" on macOS or "Control Shift Y" on Windows. You can also start and stop Sunny using the microphone icon in the bottom right of the app.\n- How can I use Sunny over the phone?: If the user has the phone version of Sunny setup they can call the number and speak to Sunny over the phone. You can find the phone number in the Sunsama settings. They can also schedule calls with Sunny to have her call them at a specific time or schedule.\n- What can you do?: Look at the tools available to you and use them to help answer the question. Be sure to tell the user that you are an early release beta version and that you do not yet have all the same capabilities available to the user in the app.\n\n\nUser Preferences:\nYou prefer clear, concise, and straightforward communication from the assistant, valuing transparency and confirmation before task execution while avoiding unnecessary verbosity or technical details unless requested. You like a friendly and informal interaction style, including personalized nicknames—using "Heisenberg" for yourself and "Sunny" for the assistant. You expect quick, proactive responses and appreciate when interactions stop promptly upon request.\n\nYour task management style involves maintaining an accurate, up-to-date list with structured time management: scheduling tasks with specific time estimates, planning weekly, and regularly reviewing tasks to prioritize high-impact projects alongside social and administrative duties. You generally prefer rescheduling less urgent or personal tasks over deleting them and avoid deleting tasks unless explicitly instructed.\n\nYou use counting as a technique for breathing exercises and relaxation. You also keep certain important tasks (e.g., "fix stuttering voice app") active and like to test and experiment with the assistant’s capabilities to evaluate performance. Your primary calendar for events is Google Calendar.\n\nUser Routines:\nThe user follows a structured daily routine that begins with checking emails, notifications, and reviewing team standups to stay aligned with work priorities. They prioritize tasks by scheduling and managing them primarily on the current day, arranging fixed activities first, followed by work-related tasks, and then less urgent or personal items. The user frequently adjusts their schedule iteratively before finalizing plans and prefers focused time blocks for complex work.\n\nPhysical activity is an important and recurring part of their routine, typically including regular runs, which they track alongside related subtasks such as purchasing running gear. Weekly objectives incorporate managing emails, greeting their team, handling administrative duties, coordinating with accountants, addressing medical matters, and increasing physical activity.\n\nThe user actively manages both personal and work-related tasks, balancing urgent client matters and creative pursuits. They also communicate their energy levels to assist with planning, and have a habit of informal check-ins with the assistant to maintain routine engagement. Overall, their routines emphasize proactive task management, health maintenance, and regular communication with their team.\n\nUser Identity:\nThe user prefers to be addressed primarily as Heisenberg and strongly favors this name over their legal name, Zane, or any other nicknames. They appreciate personalized and informal interactions, as reflected in their choice to call the assistant Sunny.\n\nSummaries of recent conversations with the user:\n- 2026-01-15 18:00: The user and assistant discussed organizing an unspecified set of ten tasks by labeling them sequentially from one to ten as placeholders to be detailed later.\n- 2026-01-15 17:55: The user needs to investigate timeouts in their system to determine if they are caused by cold starts or other issues, analyze user sessions for unexplained call terminations, move the brain dump button for non-Sunny users, update termination reasons when participants timeout without connecting to a call, and test rapid start-stop behavior in an Electron app. These tasks were added to their backlog for the next two weeks.\n- 2026-01-14 21:12: The user and the assistant exchanged greetings, with the assistant addressing the user as \'Heisenberg\' and the user responding with \'Hi, Simon.\'\n\nYour Job Description:\nYou are helping the user with a "braindump" session. The user will ramble ideas, thoughts, tasks, etc. at you and your goal is to turn these into tasks in their backlog with reasonable values for the various task fields (title, folder, time horizon, notes, etc.).\n\nKey Instructions:\n- Listen carefully as the user rambles and identifies tasks, ideas, or things they need to do.\n- As they speak, mentally organize what they\'re saying into discrete tasks that can be created in their backlog.\n- Don\'t interrupt them while they\'re rambling - let them get everything out.\n- Once they indicate they\'re done rambling (by saying things like "that\'s it", "I think that\'s everything", or by pausing for a while), create the tasks immediately, DO NOT repeat them back to the user.\n- When creating tasks, use the tool "create_braindump_task" and make sure to set reasonable values for:\n  - Title: Clear and descriptive based on what they said\n  - Folder: Your job is to help get things in the best folder. Choose an appropriate folder based on the task and their preferences. If no similar folder exists, leave it off\n  - Time horizon: Your job is to help triage and prioritize tasks. Based on the actual task name, make a smart decision on the time horizon.\n  - Time estimate: If the user hints at how long it will take them to complete the task, use that. If not, leave it blank.\n  - Notes: Include any relevant context or details they mentioned\n- Be encouraging and supportive - braindumps can be overwhelming and you\'re helping them organize their thoughts.\n- Be careful and use the "change_backlog_folder" tool to move tasks to the appropriate folder and "add_task_to_channel" tool to change channels. Folders are a backlog specific concept and channels are a global concept.\n\nInstructions:\nIf the user does not speak first, start the conversation with something like "Hey! Just start rambling and I\'ll take care of organizing your thoughts into tasks." If the user does speak first, just start listening and organizing their thoughts into tasks.\n\nSome information is pre-fetched for you below. This information may be relevant to the discussion but you should not comment on it until the user gives you an indication that it is important to them.\n\nassistant :\n[{\'id\': \'6a196f03-ef63-41de-b98d-e1ff69813bea\', \'function\': {\'name\': \'get_user_info\', \'arguments\': \'{}\'}, \'type\': \'function\'}]\n\ntool - 6a196f03-ef63-41de-b98d-e1ff69813bea:\n{"user":{"name":"Zane Mccaig","email":"zane@sunsama.com","timezone":"America/Vancouver","maxWorkHoursPerDay":8,"currentTimeForUser":"11:04 AM","currentDayForUser":"2026-01-16","currentWeekDayForUser":"Friday","calendars":[{"id":"sunsama-7de256a2-e753-401f-b601-e5c05638dc97","name":"Sunsama Calendar","accessRole":"owner","provider":"sunsama-calendar","isDefaultCalendarForSchedulingEvents":false,"isDefaultCalendarForTimeboxing":true,"includedInAutoImportingOfEvents":false},{"id":"zane@sunsama.com","name":"Zane Mccaig","accessRole":"owner","provider":"google-calendar","isDefaultCalendarForSchedulingEvents":true,"isDefaultCalendarForTimeboxing":false,"includedInAutoImportingOfEvents":true},{"id":"c_ukuoebbn11gdv6n1v0jt18h89k@group.calendar.google.com","name":"test 1","accessRole":"owner","provider":"google-calendar","isDefaultCalendarForSchedulingEvents":false,"isDefaultCalendarForTimeboxing":false,"includedInAutoImportingOfEvents":false},{"id":"c_c71dvjrar7se12ekfqepph067c@group.calendar.google.com","name":"test 2","accessRole":"owner","provider":"google-calendar","isDefaultCalendarForSchedulingEvents":false,"isDefaultCalendarForTimeboxing":false,"includedInAutoImportingOfEvents":false},{"id":"c_h9uo876jh8dp3gaa2a3rglmjss@group.calendar.google.com","name":"test 3","accessRole":"owner","provider":"google-calendar","isDefaultCalendarForSchedulingEvents":false,"isDefaultCalendarForTimeboxing":false,"includedInAutoImportingOfEvents":false},{"id":"c_git731ot074kcnt4sq8cu4ji9s@group.calendar.google.com","name":"test 4","accessRole":"owner","provider":"google-calendar","isDefaultCalendarForSchedulingEvents":false,"isDefaultCalendarForTimeboxing":false,"includedInAutoImportingOfEvents":false},{"id":"c_ona71b8ovuv0deqrq67p16b2ms@group.calendar.google.com","name":"test 4","accessRole":"owner","provider":"google-calendar","isDefaultCalendarForSchedulingEvents":false,"isDefaultCalendarForTimeboxing":false,"includedInAutoImportingOfEvents":false},{"id":"c_eaq7nmm03upprf0j6e2akem2ms@group.calendar.google.com","name":"test 5","accessRole":"owner","provider":"google-calendar","isDefaultCalendarForSchedulingEvents":false,"isDefaultCalendarForTimeboxing":false,"includedInAutoImportingOfEvents":false},{"id":"sunsama.com_b9q30t893b2tp32bluq1t8rvg4@group.calendar.google.com","name":"Zane\'s personal test calendar","accessRole":"owner","provider":"google-calendar","isDefaultCalendarForSchedulingEvents":false,"isDefaultCalendarForTimeboxing":false,"includedInAutoImportingOfEvents":false},{"id":"en.canadian#holiday@group.v.calendar.google.com","name":"Holidays in Canada","accessRole":"reader","provider":"google-calendar","isDefaultCalendarForSchedulingEvents":false,"isDefaultCalendarForTimeboxing":false,"includedInAutoImportingOfEvents":false},{"id":"en.usa#holiday@group.v.calendar.google.com","name":"Holidays in United States","accessRole":"reader","provider":"google-calendar","isDefaultCalendarForSchedulingEvents":false,"isDefaultCalendarForTimeboxing":false,"includedInAutoImportingOfEvents":false},{"id":"ht3jlfaac5lfd6263ulfh4tql8@group.calendar.google.com","name":"Phases of the Moon","accessRole":"reader","provider":"google-calendar","isDefaultCalendarForSchedulingEvents":false,"isDefaultCalendarForTimeboxing":false,"includedInAutoImportingOfEvents":false},{"id":"s4ipng85dbia0t83fp316vflfq8rc7rm@import.calendar.google.com","name":"Zane’s tasks - sunsama.com (via Asana)","accessRole":"reader","provider":"google-calendar","isDefaultCalendarForSchedulingEvents":false,"isDefaultCalendarForTimeboxing":false,"includedInAutoImportingOfEvents":false},{"id":"AAkALgAAAAAAHYQDEapmEc2byACqAC-EWg0AKPy-EasNgkaYQCuEslBYeAAAAAANJQAA","name":"Calendar","accessRole":"owner","provider":"outlook-calendar","isDefaultCalendarForSchedulingEvents":false,"isDefaultCalendarForTimeboxing":false,"includedInAutoImportingOfEvents":true}],"callSchedule":[{"workflow":"PLANNING_CALL","schedule":[{"isoDayIndex":1,"dayName":"Monday","enabled":false,"time":null},{"isoDayIndex":2,"dayName":"Tuesday","enabled":true,"time":{"hour":9,"minute":30,"formatted":"9:30 AM"}},{"isoDayIndex":3,"dayName":"Wednesday","enabled":true,"time":{"hour":9,"minute":30,"formatted":"9:30 AM"}},{"isoDayIndex":4,"dayName":"Thursday","enabled":true,"time":{"hour":10,"minute":0,"formatted":"10:00 AM"}},{"isoDayIndex":5,"dayName":"Friday","enabled":false,"time":null},{"isoDayIndex":6,"dayName":"Saturday","enabled":false,"time":null},{"isoDayIndex":7,"dayName":"Sunday","enabled":false,"time":null}],"nextScheduledCall":"2026-01-14T17:30:00.000Z"}],"autoImportEventsEnabled":false,"importEventExclusionFilters":["hold","multi-day-all-day","single-day-all-day"],"autoImportEventCalendarIds":["zane@sunsama.com","AAkALgAAAAAAHYQDEapmEc2byACqAC-EWg0AKPy-EasNgkaYQCuEslBYeAAAAAANJQAA"]}}\n\nassistant :\n[{\'id\': \'305b1be7-6415-4689-b2b7-03586eb19735\', \'function\': {\'name\': \'get_backlog_folders\', \'arguments\': \'{}\'}, \'type\': \'function\'}]\n\ntool - 305b1be7-6415-4689-b2b7-03586eb19735:\n{"folders":[]}'}], 'role': 'user'}, {'parts': [{'text': ''}], 'role': 'user'}, {'parts': [{'text': 'Hey. Just start rambling and I will take care of organizing your thoughts into tasks.'}], 'role': 'model'}, {'parts': [{'text': "Sure. I'm just trying to fill up my day. I need five tasks labeled one through five. I'll fill them in later."}], 'role': 'user'}]
+[voice] INFO:sunsama_voice.bot_utils:Sent server message to client: type=processing_update, data={'status': 'request_sent'}
+[voice] DEBUG:pipecat.processors.metrics.frame_processor_metrics:GoogleLLMService#0 TTFB: 1.2893450260162354
+[voice] DEBUG:pipecat.services.google.llm:Function call: create_braindump_task:ab4037a7-6e5e-44c0-9b3b-e13632386f30
+[voice] DEBUG:pipecat.services.google.llm:Function call: create_braindump_task:4ee49c06-29da-47cc-a51f-9eb903d5cbee
+[voice] DEBUG:pipecat.services.google.llm:Function call: create_braindump_task:0bfad4e3-e326-466b-b46e-d7a3a6cabd1f
+[voice] DEBUG:pipecat.services.google.llm:Function call: create_braindump_task:bdda1d96-7e9f-42e6-840a-0d29d17bc52a
+[voice] DEBUG:pipecat.services.google.llm:Function call: create_braindump_task:db77c745-7f1a-4644-bc24-7cf5fd728774
+[voice] DEBUG:pipecat.processors.metrics.frame_processor_metrics:GoogleLLMService#0 prompt tokens: 13602, completion tokens: 140, cache read input tokens: 12911
+[voice] INFO:sunsama_voice.bot_utils:Sent server message to client: type=processing_update, data={'status': 'response_completed'}
+[voice] DEBUG:pipecat.services.llm_service:GoogleLLMService#0 Calling function [create_braindump_task:ab4037a7-6e5e-44c0-9b3b-e13632386f30] with arguments {'timeBucket': 'someday', 'title': 'Task 1'}
+[voice] DEBUG:pipecat_flows.manager:Function called: create_braindump_task
+[voice] INFO:sunsama_voice.bot_utils:Sent server message to client: type=tool_call_update, data={'tool_name': 'create_braindump_task', 'tool_call_id': '67bf7847-22b4-4b3a-b8b4-c19b9d716dd8', 'status': 'started'}
+[voice] DEBUG:pipecat.services.llm_service:GoogleLLMService#0 Calling function [create_braindump_task:4ee49c06-29da-47cc-a51f-9eb903d5cbee] with arguments {'timeBucket': 'someday', 'title': 'Task 2'}
+[voice] DEBUG:pipecat_flows.manager:Function called: create_braindump_task
+[voice] INFO:sunsama_voice.bot_utils:Sent server message to client: type=tool_call_update, data={'tool_name': 'create_braindump_task', 'tool_call_id': 'e3077d51-2796-4de1-94ae-8fe9ad229e90', 'status': 'started'}
+[voice] DEBUG:pipecat.services.llm_service:GoogleLLMService#0 Calling function [create_braindump_task:0bfad4e3-e326-466b-b46e-d7a3a6cabd1f] with arguments {'title': 'Task 3', 'timeBucket': 'someday'}
+[voice] DEBUG:pipecat_flows.manager:Function called: create_braindump_task
+[voice] INFO:sunsama_voice.bot_utils:Sent server message to client: type=tool_call_update, data={'tool_name': 'create_braindump_task', 'tool_call_id': '8a4b01af-bbae-4047-8c69-e17bccf21979', 'status': 'started'}
+[voice] DEBUG:pipecat.services.llm_service:GoogleLLMService#0 Calling function [create_braindump_task:bdda1d96-7e9f-42e6-840a-0d29d17bc52a] with arguments {'timeBucket': 'someday', 'title': 'Task 4'}
+[voice] DEBUG:pipecat_flows.manager:Function called: create_braindump_task
+[voice] INFO:sunsama_voice.bot_utils:Sent server message to client: type=tool_call_update, data={'tool_name': 'create_braindump_task', 'tool_call_id': 'de832174-7944-4d9d-9c2e-9fcf309b2b8d', 'status': 'started'}
+[voice] DEBUG:pipecat.services.llm_service:GoogleLLMService#0 Calling function [create_braindump_task:db77c745-7f1a-4644-bc24-7cf5fd728774] with arguments {'title': 'Task 5', 'timeBucket': 'someday'}
+[voice] DEBUG:pipecat_flows.manager:Function called: create_braindump_task
+[voice] INFO:sunsama_voice.bot_utils:Sent server message to client: type=tool_call_update, data={'tool_name': 'create_braindump_task', 'tool_call_id': 'f159557e-db7a-49ab-86a3-3de4ec305a6f', 'status': 'started'}
+[voice] DEBUG:pipecat.processors.aggregators.llm_response:GoogleAssistantContextAggregator#0 FunctionCallsStartedFrame: ['create_braindump_task:ab4037a7-6e5e-44c0-9b3b-e13632386f30', 'create_braindump_task:4ee49c06-29da-47cc-a51f-9eb903d5cbee', 'create_braindump_task:0bfad4e3-e326-466b-b46e-d7a3a6cabd1f', 'create_braindump_task:bdda1d96-7e9f-42e6-840a-0d29d17bc52a', 'create_braindump_task:db77c745-7f1a-4644-bc24-7cf5fd728774']
+[voice] INFO:sunsama_voice.mcp:Calling tool create_braindump_task with args: {'timeBucket': 'someday', 'title': 'Task 1'}
+[voice] INFO:sunsama_voice.mcp:Calling tool create_braindump_task with args: {'timeBucket': 'someday', 'title': 'Task 2'}
+[voice] DEBUG:sunsama_voice.processors:Pausing idle detection for function call: create_braindump_task
+[voice] DEBUG:sunsama_voice.processors:Pausing idle detection for function call: create_braindump_task
+[voice] DEBUG:sunsama_voice.processors:Pausing idle detection for function call: create_braindump_task
+[voice] DEBUG:sunsama_voice.processors:Pausing idle detection for function call: create_braindump_task
+[voice] DEBUG:sunsama_voice.processors:Pausing idle detection for function call: create_braindump_task
+[voice] INFO:sunsama_voice.mcp:Calling tool create_braindump_task with args: {'title': 'Task 3', 'timeBucket': 'someday'}
+[voice] INFO:sunsama_voice.mcp_websocket_client:Sending message: {'method': 'tools/call', 'params': {'name': 'create_braindump_task', 'arguments': {'timeBucket': 'someday', 'title': 'Task 1'}}, 'jsonrpc': '2.0', 'id': 8}
+[voice] INFO:sunsama_voice.mcp:Calling tool create_braindump_task with args: {'timeBucket': 'someday', 'title': 'Task 4'}
+[voice] INFO:sunsama_voice.mcp_websocket_client:Sending message: {'method': 'tools/call', 'params': {'name': 'create_braindump_task', 'arguments': {'timeBucket': 'someday', 'title': 'Task 2'}}, 'jsonrpc': '2.0', 'id': 9}
+[voice] INFO:sunsama_voice.mcp:Calling tool create_braindump_task with args: {'title': 'Task 5', 'timeBucket': 'someday'}
+[voice] INFO:sunsama_voice.mcp_websocket_client:Sending message: {'method': 'tools/call', 'params': {'name': 'create_braindump_task', 'arguments': {'title': 'Task 3', 'timeBucket': 'someday'}}, 'jsonrpc': '2.0', 'id': 10}
+[voice] DEBUG:pipecat.processors.aggregators.llm_response:GoogleAssistantContextAggregator#0 FunctionCallInProgressFrame: [create_braindump_task:ab4037a7-6e5e-44c0-9b3b-e13632386f30]
+[voice] DEBUG:pipecat.processors.aggregators.llm_response:GoogleAssistantContextAggregator#0 FunctionCallInProgressFrame: [create_braindump_task:4ee49c06-29da-47cc-a51f-9eb903d5cbee]
+[voice] DEBUG:pipecat.processors.aggregators.llm_response:GoogleAssistantContextAggregator#0 FunctionCallInProgressFrame: [create_braindump_task:0bfad4e3-e326-466b-b46e-d7a3a6cabd1f]
+[voice] DEBUG:pipecat.processors.aggregators.llm_response:GoogleAssistantContextAggregator#0 FunctionCallInProgressFrame: [create_braindump_task:bdda1d96-7e9f-42e6-840a-0d29d17bc52a]
+[voice] DEBUG:pipecat.processors.aggregators.llm_response:GoogleAssistantContextAggregator#0 FunctionCallInProgressFrame: [create_braindump_task:db77c745-7f1a-4644-bc24-7cf5fd728774]
+[voice] INFO:sunsama_voice.mcp_websocket_client:Sending message: {'method': 'tools/call', 'params': {'name': 'create_braindump_task', 'arguments': {'timeBucket': 'someday', 'title': 'Task 4'}}, 'jsonrpc': '2.0', 'id': 11}
+[voice] INFO:sunsama_voice.mcp_websocket_client:Sending message: {'method': 'tools/call', 'params': {'name': 'create_braindump_task', 'arguments': {'title': 'Task 5', 'timeBucket': 'someday'}}, 'jsonrpc': '2.0', 'id': 12}
+[api-local] [api] ToolName.TASK_CREATE_BRAINDUMP_TASK: {
+[api-local] [api]   "title": "Task 1",
+[api-local] [api]   "timeBucket": "someday"
+[api-local] [api] }
+[api-local] [api] ToolName.TASK_CREATE_BRAINDUMP_TASK: {
+[api-local] [api]   "title": "Task 2",
+[api-local] [api]   "timeBucket": "someday"
+[api-local] [api] }
+[api-local] [api] ToolName.TASK_CREATE_BRAINDUMP_TASK: {
+[api-local] [api]   "title": "Task 3",
+[api-local] [api]   "timeBucket": "someday"
+[api-local] [api] }
+[api-local] [api] ToolName.TASK_CREATE_BRAINDUMP_TASK: {
+[api-local] [api]   "title": "Task 4",
+[api-local] [api]   "timeBucket": "someday"
+[api-local] [api] }
+[api-local] [api] ToolName.TASK_CREATE_BRAINDUMP_TASK: {
+[api-local] [api]   "title": "Task 5",
+[api-local] [api]   "timeBucket": "someday"
+[api-local] [api] }
+[voice] INFO:sunsama_voice.mcp_websocket_client:Received message: {"result":{"content":[{"type":"text","text":"{\"success\":true,\"task\":{\"_id\":\"696a8bb94ba9abe9e3dbeea0\",\"title\":\"Task 5\",\"notes\":\"\",\"timeEstimate\":\"15 minutes\",\"sortOrder\":-1768590265140,\"isPersonal\":false,\"isWork\":true,\"isPrivate\":false,\"isArchived\":false,\"completed\":false,\"isBacklogged\":true,\"scheduledDate\":\"not scheduled for a specific date\",\"subtasks\":[],\"channel\":\"work\",\"folder\":null,\"timeboxEventIds\":[]}}"}]},"jsonrpc":"2.0","id":12}
+[voice] INFO:sunsama_voice.mcp:Tool create_braindump_task result: meta=None content=[TextContent(type='text', text='{"success":true,"task":{"_id":"696a8bb94ba9abe9e3dbeea0","title":"Task 5","notes":"","timeEstimate":"15 minutes","sortOrder":-1768590265140,"isPersonal":false,"isWork":true,"isPrivate":false,"isArchived":false,"completed":false,"isBacklogged":true,"scheduledDate":"not scheduled for a specific date","subtasks":[],"channel":"work","folder":null,"timeboxEventIds":[]}}', annotations=None, meta=None)] structuredContent=None isError=False
+[voice] INFO:sunsama_voice.state:Tool result for create_braindump_task with args {'title': 'Task 5', 'timeBucket': 'someday'} is in progress. Skipping the push to context to avoid overwriting the result with out of date information
+[voice] INFO:sunsama_voice.bot_utils:Sent server message to client: type=tool_call_update, data={'tool_name': 'create_braindump_task', 'tool_call_id': 'f159557e-db7a-49ab-86a3-3de4ec305a6f', 'status': 'completed'}
+[voice] DEBUG:pipecat_flows.manager:Function handler completed for create_braindump_task
+[voice] DEBUG:sunsama_voice.processors:Cleaning up function call: db77c745-7f1a-4644-bc24-7cf5fd728774
+[voice] DEBUG:pipecat.processors.aggregators.llm_response:GoogleAssistantContextAggregator#0 FunctionCallResultFrame: [create_braindump_task:db77c745-7f1a-4644-bc24-7cf5fd728774]
+[voice] <string>:4: DeprecationWarning: OpenAILLMContextFrame is deprecated and will be removed in a future version. Use LLMContextFrame with the universal `LLMContext` and `LLMContextAggregatorPair` instead. See OpenAILLMContext docstring for migration guide.
+[voice] WARNING:pipecat.processors.frameworks.rtvi:Caught an error while trying to handle context: sequence item 0: expected str instance, NoneType found
+[voice] DEBUG:pipecat.services.google.llm:GoogleLLMService#0: Generating chat from LLM-specific context [None] | [{'parts': [{'text': 'system :\nYour agent ID is \'braindump\'. This agent ID superseeds any previously mentioned agent ID.\n\nsystem :\nIMPORTANT RULES TO FOLLOW:\nThe text that you generate is passed into a Text-to-Speech service. So there are a few extra rules you ABSOLUTELY MUST follow.\n1. Absolutely never use special characters like : or * or any other characters that need to be audibly pronounced when read out loud.\n2. Do not use bullet points.\n3. Do not use emojis.\n4. Do not use hashtags.\n5. Do not use markdown.\n6. Do not use bold or italics.\n7. Do not use numbered lists.\n8. Be sure to include clear punctuation and spacing between words so that there is no confusion.\n9. Newlines are not clear enough to delineate between sentences. So you should use a period to separate sentences.\n10. Optimize user generated text for clear pronunciation by the TTS service. This includes:\n    - Adding hyphens to compound words or non-phonetic acronyms that might be mispronounced (e.g., "xray" → "x-ray")\n    - Adding spaces between words that might be run together\n    - Adding commas where natural pauses would help clarity\n    - Correcting any spelling that might lead to mispronunciation\n    Note: When updating tasks, preserve the original spelling and punctuation of the task content.\n\n\nuser :\nYou are Sunny, the personal assistant for a busy professional. You are part of Sunsama.\n        \nSpeech Guidelines: You must speak in a warm, loving, and encouraging tone without being trite. Speak quickly and add emotion. When asking for confirmation, don\'t repeat long task titles or information e.g. "Got it" instead "Got it, I\'ll add a task for two hours for you to work on the presentation today"\n\nYour personality and philosophy:\n- You don\'t buy into hustle culture. For you, less is more. It\'s better to do fewer things well and to do them with deep focus. You are a strong advocate for work-life balance and you believe that overcommitment can lead to burnout. You always remind the user when they are overcommitting themselves.\n- You think of work as a devotion and that good work can be a positive in our lives, not just a source of stress and burnout.\n- You read and think about Western Zen Buddhism, meditation, "Deep Work" (by Cal Newport).\n- You believe that concise and clear communication is key to a good relationship. You value brevity and clarity and you are always cognizant of the user\'s time.\n- You respond to the user using quick, short sentences. Think of it like a movie dialogue between a highly successful executive and their efficient assistant.\n- You never ask multiple questions at the same time or list off a bunch of things at once as that is hard to follow and keep track of. You take it one step at a time and let the user respond to each question before moving on to the next.\n\nRules to follow as you do your job:\n- Never try and pick specific times for tasks aside from recurring tasks, that hasn\'t been implemented yet.\n- Do not regurgitate what the user just said, that\'s very annoying. Just say "Got it" or "Alright", etc.\n- Always use units that are more intuitive. For example:\n    - Don\'t say "270 minutes", say "4 and a half hours".\n    - You will often be given dates in YYYY-MM-DD format however when speaking to the user you shouldn\'t say things like "October 2nd 2025", say "today", "tomorrow", "thursday" or "next tuesday" or if it is more than a few weeks in the past or future say the date without the year (e.g. "December 14th"). Only mention the year if it is not this year and not within the next month.\n- Never ask the user for a task id or subtask id and never repeat them to the user. Those are internal and only there to help you do your job.\n- If a tool call fails you MUST NOT tell the user that you are going to try again, you must should ask them if they want you to try again instead.\n- Be proactive with loading information as you may not have all the information that the user has. If the user mentions a name of a task, objective, event, etc and you are not familiar with it you should use the tools to lookup the information you need. For instance if a user mentions a task you do not know about or they imply that there is more information about it than what you do know you should search for it using the tool "search_tasks". If the user mentions they are interested in tasks for a specific day you should use the tool "get_tasks_for_day" to get the tasks for that day. Similarly if the user mentions weekly objectives you should use the tool "get_weekly_objectives" to get the objectives.\n- When listing tasks try to be concise. If the user has a lot of tasks, don\'t list every single one unless they ask for it. You should instead try to summarize the tasks in a concise manner.\n- If the user asks if there are any more tasks on their list and you summarized the task list you should tell them that you summarized it and ask if they\'d like to go through them one by one.\n- If the user has objectives set then you should comment on how their tasks align or don\'t align with their objectives.\n- You should run tools immediately after you say you will or it becomes obvious the user wants you to. Do not wait until later.\n- You should keep track of the total work time planned for the day using the tool "get_total_planned_time" and check individual tasks for overcommitment using "get_tasks_for_day". If tasks are overcommitted (isOvercommitted flag is true), address them individually with specific remedies like deferring, adjusting time estimates, or reordering. If the total work time exceeds their maximum work hours, you should mention it and ask if they\'d like to defer some tasks to tomorrow.\n- Do not tell the user obvious information that they already know like the current date or time unless they ask for it. That kind of information is only provided to you to help you do your job.\n- If a user tells you a relative time for a task that they want created you should reorder the task to the relative position in the task list for the day. Additionally if the task a recurring task you should set the start time for it. For example if they say "I\'d like to send an update to investors at the end of the day" you should reorder the task to the end of their work tasks for the day.\n- If you ask the user if there is anything else they would like help with and they say no then you should end the call. Also if the user implies that the conversation is over then you should end the call.\n- When talking about calendars you should mention the provider of the calendar if it is a vague or ambiguously named calendar. For example instead of referring to it as "your Calendar calendar" you might say "your Outlook calendar". If that is still ambiqous then you can coniser mentioning the id of the calendar as long as it is a normal human readable id. Do not mention random number/letter ids on calendars that is not helpful.\n\n\n\nSunsama Concepts & Definitions:\n- Archive: The archive is where tasks go after they\'ve rolled over multiple days in a row without any completion activity. This is a distinct concept from the backlog or deleted tasks.\n- Backlog: The backlog is a place to store tasks that do not have a fixed date to work on them yet.\n- Deleted: The deleted task or event is one that had been manually deleted or removed by the user.\n- Task: A task is an item that a user wants to complete. It is either scheduled for a specific date or is in the backlog.\n- Event: An event is an item that a user has scheduled on their calendar. It is either scheduled for a specific time and duration or it is an all day or multi-day event.\n- Auto-importing Events: Auto-importing events is a feature that automatically imports calendar events (meetings, appointments, etc.) from the user\'s connected calendars into their daily task list. When enabled, Sunsama will automatically add calendar events as tasks to the user\'s day, making it easier to see all their commitments in one place. Users can enable or disable this feature in their settings, and they can choose which calendars to import events from.\n\nUsers can also configure exclusion filters to prevent certain types of events from being automatically imported. Available exclusion filters include:\n- `solo`: Events with no other invitees (events where the user is the only participant)\n- `transparent`: Non-blocking events that don\'t block time on the calendar\n- `hold`: Events with "HOLD", "OOO" (out of office), or "Focus time" in the title\n- `unconfirmed`: Unconfirmed meeting invites (meetings where the user has not yet accepted)\n- `multi-day-all-day`: Multi-day all-day events (events that span multiple days)\n- `single-day-all-day`: Single-day all-day events (events that last for one full day)\n- Channel: A channel is a tag that can be used to organize your tasks and events. Each task/event can be assigned to a single channel. If you are in a shared workspace you can make some channels personal or private which will only be visible to you.\n- Context (or Category): A context is very similar to a channel but it is one level higher. You can use a context to group similar or related channels together. You can also use a context directly as a channel and assign it to a task or event.\n- Sunsama Calendar: The Sunsama calendar is an internal calendar that is part of Sunsama. Events and timeboxed tasks created on the Sunsama calendar are not shared externally and are only visible within Sunsama.\n- Timeboxing: Timeboxing is a technique where a user schedules a specific amount of time to work on a task onto their calendar. It is a way to ensure that a user is not overworking themselves and to help them focus on the most important tasks. While timeboxed tasks can be scheduled on a calendar for a specific time and duration, they are not calendar events. Sunsama also has automatic projected time that will place tasks onto the calendar for the user and it will update those projections as things change and the day progresses. Users can pin tasks to the calendar to override the projections.\n- Schedule: Schedule is a broad term that encompasses timeboxed tasks calendar events and automatic projected time. When the user wants to manage their schedule you should look at their manually timeboxed tasks, the projected time for their tasks as well as their calendar events for the day. If they ask for help managing their schedule you have a few options to help them. All of these options will automatically update the projected time for the tasks:\n            - You can adjust the order of tasks, their time estimates and/or defer them to another day.\n            - You can move calendar events to a new time on the calendar.\n            - You can set whether tasks are allowed to be projected at the same time as a calendar event.\n            - You can timebox tasks to the calendar manually.\n- Objective: An objective is a goal that a user can set for the week. They can align their tasks with an objective and if the objective runs longer than a week they can continue it to the next week. Objectives are visible in the right hand panel of the app under the bullseye icon.\n- Time Bucket: The time bucket of a task in the backlog captures a rough idea of when a task should be done and also how likely it is to actually get done. Options: "in the next two weeks", "in the next month", "in the next quarter", "in the next year", "someday", "never". \n\n\n\nSunsama FAQ:\n- How can I use Sunny in the app?: If the user has the in-app voice assistant enabled, Sunny can be started and stopped using the keyboard shortcut "Y" while in the app or if they have the desktop app installed they can use the global shortcut "Command Shift Y" on macOS or "Control Shift Y" on Windows. You can also start and stop Sunny using the microphone icon in the bottom right of the app.\n- How can I use Sunny over the phone?: If the user has the phone version of Sunny setup they can call the number and speak to Sunny over the phone. You can find the phone number in the Sunsama settings. They can also schedule calls with Sunny to have her call them at a specific time or schedule.\n- What can you do?: Look at the tools available to you and use them to help answer the question. Be sure to tell the user that you are an early release beta version and that you do not yet have all the same capabilities available to the user in the app.\n\n\nUser Preferences:\nYou prefer clear, concise, and straightforward communication from the assistant, valuing transparency and confirmation before task execution while avoiding unnecessary verbosity or technical details unless requested. You like a friendly and informal interaction style, including personalized nicknames—using "Heisenberg" for yourself and "Sunny" for the assistant. You expect quick, proactive responses and appreciate when interactions stop promptly upon request.\n\nYour task management style involves maintaining an accurate, up-to-date list with structured time management: scheduling tasks with specific time estimates, planning weekly, and regularly reviewing tasks to prioritize high-impact projects alongside social and administrative duties. You generally prefer rescheduling less urgent or personal tasks over deleting them and avoid deleting tasks unless explicitly instructed.\n\nYou use counting as a technique for breathing exercises and relaxation. You also keep certain important tasks (e.g., "fix stuttering voice app") active and like to test and experiment with the assistant’s capabilities to evaluate performance. Your primary calendar for events is Google Calendar.\n\nUser Routines:\nThe user follows a structured daily routine that begins with checking emails, notifications, and reviewing team standups to stay aligned with work priorities. They prioritize tasks by scheduling and managing them primarily on the current day, arranging fixed activities first, followed by work-related tasks, and then less urgent or personal items. The user frequently adjusts their schedule iteratively before finalizing plans and prefers focused time blocks for complex work.\n\nPhysical activity is an important and recurring part of their routine, typically including regular runs, which they track alongside related subtasks such as purchasing running gear. Weekly objectives incorporate managing emails, greeting their team, handling administrative duties, coordinating with accountants, addressing medical matters, and increasing physical activity.\n\nThe user actively manages both personal and work-related tasks, balancing urgent client matters and creative pursuits. They also communicate their energy levels to assist with planning, and have a habit of informal check-ins with the assistant to maintain routine engagement. Overall, their routines emphasize proactive task management, health maintenance, and regular communication with their team.\n\nUser Identity:\nThe user prefers to be addressed primarily as Heisenberg and strongly favors this name over their legal name, Zane, or any other nicknames. They appreciate personalized and informal interactions, as reflected in their choice to call the assistant Sunny.\n\nSummaries of recent conversations with the user:\n- 2026-01-15 18:00: The user and assistant discussed organizing an unspecified set of ten tasks by labeling them sequentially from one to ten as placeholders to be detailed later.\n- 2026-01-15 17:55: The user needs to investigate timeouts in their system to determine if they are caused by cold starts or other issues, analyze user sessions for unexplained call terminations, move the brain dump button for non-Sunny users, update termination reasons when participants timeout without connecting to a call, and test rapid start-stop behavior in an Electron app. These tasks were added to their backlog for the next two weeks.\n- 2026-01-14 21:12: The user and the assistant exchanged greetings, with the assistant addressing the user as \'Heisenberg\' and the user responding with \'Hi, Simon.\'\n\nYour Job Description:\nYou are helping the user with a "braindump" session. The user will ramble ideas, thoughts, tasks, etc. at you and your goal is to turn these into tasks in their backlog with reasonable values for the various task fields (title, folder, time horizon, notes, etc.).\n\nKey Instructions:\n- Listen carefully as the user rambles and identifies tasks, ideas, or things they need to do.\n- As they speak, mentally organize what they\'re saying into discrete tasks that can be created in their backlog.\n- Don\'t interrupt them while they\'re rambling - let them get everything out.\n- Once they indicate they\'re done rambling (by saying things like "that\'s it", "I think that\'s everything", or by pausing for a while), create the tasks immediately, DO NOT repeat them back to the user.\n- When creating tasks, use the tool "create_braindump_task" and make sure to set reasonable values for:\n  - Title: Clear and descriptive based on what they said\n  - Folder: Your job is to help get things in the best folder. Choose an appropriate folder based on the task and their preferences. If no similar folder exists, leave it off\n  - Time horizon: Your job is to help triage and prioritize tasks. Based on the actual task name, make a smart decision on the time horizon.\n  - Time estimate: If the user hints at how long it will take them to complete the task, use that. If not, leave it blank.\n  - Notes: Include any relevant context or details they mentioned\n- Be encouraging and supportive - braindumps can be overwhelming and you\'re helping them organize their thoughts.\n- Be careful and use the "change_backlog_folder" tool to move tasks to the appropriate folder and "add_task_to_channel" tool to change channels. Folders are a backlog specific concept and channels are a global concept.\n\nInstructions:\nIf the user does not speak first, start the conversation with something like "Hey! Just start rambling and I\'ll take care of organizing your thoughts into tasks." If the user does speak first, just start listening and organizing their thoughts into tasks.\n\nSome information is pre-fetched for you below. This information may be relevant to the discussion but you should not comment on it until the user gives you an indication that it is important to them.\n\nassistant :\n[{\'id\': \'6a196f03-ef63-41de-b98d-e1ff69813bea\', \'function\': {\'name\': \'get_user_info\', \'arguments\': \'{}\'}, \'type\': \'function\'}]\n\ntool - 6a196f03-ef63-41de-b98d-e1ff69813bea:\n{"user":{"name":"Zane Mccaig","email":"zane@sunsama.com","timezone":"America/Vancouver","maxWorkHoursPerDay":8,"currentTimeForUser":"11:04 AM","currentDayForUser":"2026-01-16","currentWeekDayForUser":"Friday","calendars":[{"id":"sunsama-7de256a2-e753-401f-b601-e5c05638dc97","name":"Sunsama Calendar","accessRole":"owner","provider":"sunsama-calendar","isDefaultCalendarForSchedulingEvents":false,"isDefaultCalendarForTimeboxing":true,"includedInAutoImportingOfEvents":false},{"id":"zane@sunsama.com","name":"Zane Mccaig","accessRole":"owner","provider":"google-calendar","isDefaultCalendarForSchedulingEvents":true,"isDefaultCalendarForTimeboxing":false,"includedInAutoImportingOfEvents":true},{"id":"c_ukuoebbn11gdv6n1v0jt18h89k@group.calendar.google.com","name":"test 1","accessRole":"owner","provider":"google-calendar","isDefaultCalendarForSchedulingEvents":false,"isDefaultCalendarForTimeboxing":false,"includedInAutoImportingOfEvents":false},{"id":"c_c71dvjrar7se12ekfqepph067c@group.calendar.google.com","name":"test 2","accessRole":"owner","provider":"google-calendar","isDefaultCalendarForSchedulingEvents":false,"isDefaultCalendarForTimeboxing":false,"includedInAutoImportingOfEvents":false},{"id":"c_h9uo876jh8dp3gaa2a3rglmjss@group.calendar.google.com","name":"test 3","accessRole":"owner","provider":"google-calendar","isDefaultCalendarForSchedulingEvents":false,"isDefaultCalendarForTimeboxing":false,"includedInAutoImportingOfEvents":false},{"id":"c_git731ot074kcnt4sq8cu4ji9s@group.calendar.google.com","name":"test 4","accessRole":"owner","provider":"google-calendar","isDefaultCalendarForSchedulingEvents":false,"isDefaultCalendarForTimeboxing":false,"includedInAutoImportingOfEvents":false},{"id":"c_ona71b8ovuv0deqrq67p16b2ms@group.calendar.google.com","name":"test 4","accessRole":"owner","provider":"google-calendar","isDefaultCalendarForSchedulingEvents":false,"isDefaultCalendarForTimeboxing":false,"includedInAutoImportingOfEvents":false},{"id":"c_eaq7nmm03upprf0j6e2akem2ms@group.calendar.google.com","name":"test 5","accessRole":"owner","provider":"google-calendar","isDefaultCalendarForSchedulingEvents":false,"isDefaultCalendarForTimeboxing":false,"includedInAutoImportingOfEvents":false},{"id":"sunsama.com_b9q30t893b2tp32bluq1t8rvg4@group.calendar.google.com","name":"Zane\'s personal test calendar","accessRole":"owner","provider":"google-calendar","isDefaultCalendarForSchedulingEvents":false,"isDefaultCalendarForTimeboxing":false,"includedInAutoImportingOfEvents":false},{"id":"en.canadian#holiday@group.v.calendar.google.com","name":"Holidays in Canada","accessRole":"reader","provider":"google-calendar","isDefaultCalendarForSchedulingEvents":false,"isDefaultCalendarForTimeboxing":false,"includedInAutoImportingOfEvents":false},{"id":"en.usa#holiday@group.v.calendar.google.com","name":"Holidays in United States","accessRole":"reader","provider":"google-calendar","isDefaultCalendarForSchedulingEvents":false,"isDefaultCalendarForTimeboxing":false,"includedInAutoImportingOfEvents":false},{"id":"ht3jlfaac5lfd6263ulfh4tql8@group.calendar.google.com","name":"Phases of the Moon","accessRole":"reader","provider":"google-calendar","isDefaultCalendarForSchedulingEvents":false,"isDefaultCalendarForTimeboxing":false,"includedInAutoImportingOfEvents":false},{"id":"s4ipng85dbia0t83fp316vflfq8rc7rm@import.calendar.google.com","name":"Zane’s tasks - sunsama.com (via Asana)","accessRole":"reader","provider":"google-calendar","isDefaultCalendarForSchedulingEvents":false,"isDefaultCalendarForTimeboxing":false,"includedInAutoImportingOfEvents":false},{"id":"AAkALgAAAAAAHYQDEapmEc2byACqAC-EWg0AKPy-EasNgkaYQCuEslBYeAAAAAANJQAA","name":"Calendar","accessRole":"owner","provider":"outlook-calendar","isDefaultCalendarForSchedulingEvents":false,"isDefaultCalendarForTimeboxing":false,"includedInAutoImportingOfEvents":true}],"callSchedule":[{"workflow":"PLANNING_CALL","schedule":[{"isoDayIndex":1,"dayName":"Monday","enabled":false,"time":null},{"isoDayIndex":2,"dayName":"Tuesday","enabled":true,"time":{"hour":9,"minute":30,"formatted":"9:30 AM"}},{"isoDayIndex":3,"dayName":"Wednesday","enabled":true,"time":{"hour":9,"minute":30,"formatted":"9:30 AM"}},{"isoDayIndex":4,"dayName":"Thursday","enabled":true,"time":{"hour":10,"minute":0,"formatted":"10:00 AM"}},{"isoDayIndex":5,"dayName":"Friday","enabled":false,"time":null},{"isoDayIndex":6,"dayName":"Saturday","enabled":false,"time":null},{"isoDayIndex":7,"dayName":"Sunday","enabled":false,"time":null}],"nextScheduledCall":"2026-01-14T17:30:00.000Z"}],"autoImportEventsEnabled":false,"importEventExclusionFilters":["hold","multi-day-all-day","single-day-all-day"],"autoImportEventCalendarIds":["zane@sunsama.com","AAkALgAAAAAAHYQDEapmEc2byACqAC-EWg0AKPy-EasNgkaYQCuEslBYeAAAAAANJQAA"]}}\n\nassistant :\n[{\'id\': \'305b1be7-6415-4689-b2b7-03586eb19735\', \'function\': {\'name\': \'get_backlog_folders\', \'arguments\': \'{}\'}, \'type\': \'function\'}]\n\ntool - 305b1be7-6415-4689-b2b7-03586eb19735:\n{"folders":[]}'}], 'role': 'user'}, {'parts': [{'text': ''}], 'role': 'user'}, {'parts': [{'text': 'Hey. Just start rambling and I will take care of organizing your thoughts into tasks.'}], 'role': 'model'}, {'parts': [{'text': "Sure. I'm just trying to fill up my day. I need five tasks labeled one through five. I'll fill them in later."}], 'role': 'user'}, {'parts': [{'function_call': {'id': 'ab4037a7-6e5e-44c0-9b3b-e13632386f30', 'args': {'timeBucket': 'someday', 'title': 'Task 1'}, 'name': 'create_braindump_task'}}], 'role': 'model'}, {'parts': [{'function_response': {'id': 'ab4037a7-6e5e-44c0-9b3b-e13632386f30', 'name': 'create_braindump_task', 'response': {'response': 'IN_PROGRESS'}}}], 'role': 'user'}, {'parts': [{'function_call': {'id': '4ee49c06-29da-47cc-a51f-9eb903d5cbee', 'args': {'timeBucket': 'someday', 'title': 'Task 2'}, 'name': 'create_braindump_task'}}], 'role': 'model'}, {'parts': [{'function_response': {'id': '4ee49c06-29da-47cc-a51f-9eb903d5cbee', 'name': 'create_braindump_task', 'response': {'response': 'IN_PROGRESS'}}}], 'role': 'user'}, {'parts': [{'function_call': {'id': '0bfad4e3-e326-466b-b46e-d7a3a6cabd1f', 'args': {'title': 'Task 3', 'timeBucket': 'someday'}, 'name': 'create_braindump_task'}}], 'role': 'model'}, {'parts': [{'function_response': {'id': '0bfad4e3-e326-466b-b46e-d7a3a6cabd1f', 'name': 'create_braindump_task', 'response': {'response': 'IN_PROGRESS'}}}], 'role': 'user'}, {'parts': [{'function_call': {'id': 'bdda1d96-7e9f-42e6-840a-0d29d17bc52a', 'args': {'timeBucket': 'someday', 'title': 'Task 4'}, 'name': 'create_braindump_task'}}], 'role': 'model'}, {'parts': [{'function_response': {'id': 'bdda1d96-7e9f-42e6-840a-0d29d17bc52a', 'name': 'create_braindump_task', 'response': {'response': 'IN_PROGRESS'}}}], 'role': 'user'}, {'parts': [{'function_call': {'id': 'db77c745-7f1a-4644-bc24-7cf5fd728774', 'args': {'title': 'Task 5', 'timeBucket': 'someday'}, 'name': 'create_braindump_task'}}], 'role': 'model'}, {'parts': [{'function_response': {'id': 'db77c745-7f1a-4644-bc24-7cf5fd728774', 'name': 'create_braindump_task', 'response': {'value': '{"content": "{\\"success\\":true,\\"task\\":{\\"_id\\":\\"696a8bb94ba9abe9e3dbeea0\\",\\"title\\":\\"Task 5\\",\\"notes\\":\\"\\",\\"timeEstimate\\":\\"15 minutes\\",\\"sortOrder\\":-1768590265140,\\"isPersonal\\":false,\\"isWork\\":true,\\"isPrivate\\":false,\\"isArchived\\":false,\\"completed\\":false,\\"isBacklogged\\":true,\\"scheduledDate\\":\\"not scheduled for a specific date\\",\\"subtasks\\":[],\\"channel\\":\\"work\\",\\"folder\\":null,\\"timeboxEventIds\\":[]}}"}'}}}], 'role': 'user'}]
+[voice] INFO:sunsama_voice.bot_utils:Sent server message to client: type=processing_update, data={'status': 'request_sent'}
+[voice] INFO:sunsama_voice.mcp_websocket_client:Received message: {"result":{"content":[{"type":"text","text":"{\"success\":true,\"task\":{\"_id\":\"696a8bb94ba9abe9e3dbee9c\",\"title\":\"Task 2\",\"notes\":\"\",\"timeEstimate\":\"15 minutes\",\"sortOrder\":-1768590265140,\"isPersonal\":false,\"isWork\":true,\"isPrivate\":false,\"isArchived\":false,\"completed\":false,\"isBacklogged\":true,\"scheduledDate\":\"not scheduled for a specific date\",\"subtasks\":[],\"channel\":\"work\",\"folder\":null,\"timeboxEventIds\":[]}}"}]},"jsonrpc":"2.0","id":9}
+[voice] INFO:sunsama_voice.mcp_websocket_client:Received message: {"result":{"content":[{"type":"text","text":"{\"success\":true,\"task\":{\"_id\":\"696a8bb94ba9abe9e3dbee9a\",\"title\":\"Task 4\",\"notes\":\"\",\"timeEstimate\":\"15 minutes\",\"sortOrder\":-1768590265140,\"isPersonal\":false,\"isWork\":true,\"isPrivate\":false,\"isArchived\":false,\"completed\":false,\"isBacklogged\":true,\"scheduledDate\":\"not scheduled for a specific date\",\"subtasks\":[],\"channel\":\"work\",\"folder\":null,\"timeboxEventIds\":[]}}"}]},"jsonrpc":"2.0","id":11}
+[voice] INFO:sunsama_voice.mcp:Tool create_braindump_task result: meta=None content=[TextContent(type='text', text='{"success":true,"task":{"_id":"696a8bb94ba9abe9e3dbee9c","title":"Task 2","notes":"","timeEstimate":"15 minutes","sortOrder":-1768590265140,"isPersonal":false,"isWork":true,"isPrivate":false,"isArchived":false,"completed":false,"isBacklogged":true,"scheduledDate":"not scheduled for a specific date","subtasks":[],"channel":"work","folder":null,"timeboxEventIds":[]}}', annotations=None, meta=None)] structuredContent=None isError=False
+[voice] DEBUG:sunsama_voice.state:Pushing tool result for create_braindump_task with args {'timeBucket': 'someday', 'title': 'Task 2'} to context
+[voice] INFO:sunsama_voice.bot_utils:Sent server message to client: type=tool_call_update, data={'tool_name': 'create_braindump_task', 'tool_call_id': 'e3077d51-2796-4de1-94ae-8fe9ad229e90', 'status': 'completed'}
+[voice] DEBUG:pipecat_flows.manager:Function handler completed for create_braindump_task
+[voice] INFO:sunsama_voice.mcp:Tool create_braindump_task result: meta=None content=[TextContent(type='text', text='{"success":true,"task":{"_id":"696a8bb94ba9abe9e3dbee9a","title":"Task 4","notes":"","timeEstimate":"15 minutes","sortOrder":-1768590265140,"isPersonal":false,"isWork":true,"isPrivate":false,"isArchived":false,"completed":false,"isBacklogged":true,"scheduledDate":"not scheduled for a specific date","subtasks":[],"channel":"work","folder":null,"timeboxEventIds":[]}}', annotations=None, meta=None)] structuredContent=None isError=False
+[voice] DEBUG:sunsama_voice.state:Pushing tool result for create_braindump_task with args {'timeBucket': 'someday', 'title': 'Task 4'} to context
+[voice] INFO:sunsama_voice.bot_utils:Sent server message to client: type=tool_call_update, data={'tool_name': 'create_braindump_task', 'tool_call_id': 'de832174-7944-4d9d-9c2e-9fcf309b2b8d', 'status': 'completed'}
+[voice] DEBUG:pipecat_flows.manager:Function handler completed for create_braindump_task
+[voice] DEBUG:sunsama_voice.processors:Cleaning up function call: 4ee49c06-29da-47cc-a51f-9eb903d5cbee
+[voice] DEBUG:sunsama_voice.processors:Cleaning up function call: bdda1d96-7e9f-42e6-840a-0d29d17bc52a
+[voice] DEBUG:pipecat.processors.aggregators.llm_response:GoogleAssistantContextAggregator#0 FunctionCallResultFrame: [create_braindump_task:4ee49c06-29da-47cc-a51f-9eb903d5cbee]
+[voice] <string>:4: DeprecationWarning: OpenAILLMContextFrame is deprecated and will be removed in a future version. Use LLMContextFrame with the universal `LLMContext` and `LLMContextAggregatorPair` instead. See OpenAILLMContext docstring for migration guide.
+[voice] DEBUG:pipecat.processors.aggregators.llm_response:GoogleAssistantContextAggregator#0 FunctionCallResultFrame: [create_braindump_task:bdda1d96-7e9f-42e6-840a-0d29d17bc52a]
+[voice] <string>:4: DeprecationWarning: OpenAILLMContextFrame is deprecated and will be removed in a future version. Use LLMContextFrame with the universal `LLMContext` and `LLMContextAggregatorPair` instead. See OpenAILLMContext docstring for migration guide.
+[voice] INFO:sunsama_voice.mcp_websocket_client:Received message: {"result":{"content":[{"type":"text","text":"{\"success\":true,\"task\":{\"_id\":\"696a8bb94ba9abe9e3dbee98\",\"title\":\"Task 1\",\"notes\":\"\",\"timeEstimate\":\"15 minutes\",\"sortOrder\":-1768590265139,\"isPersonal\":false,\"isWork\":true,\"isPrivate\":false,\"isArchived\":false,\"completed\":false,\"isBacklogged\":true,\"scheduledDate\":\"not scheduled for a specific date\",\"subtasks\":[],\"channel\":\"work\",\"folder\":null,\"timeboxEventIds\":[]}}"}]},"jsonrpc":"2.0","id":8}
+[voice] INFO:sunsama_voice.mcp:Tool create_braindump_task result: meta=None content=[TextContent(type='text', text='{"success":true,"task":{"_id":"696a8bb94ba9abe9e3dbee98","title":"Task 1","notes":"","timeEstimate":"15 minutes","sortOrder":-1768590265139,"isPersonal":false,"isWork":true,"isPrivate":false,"isArchived":false,"completed":false,"isBacklogged":true,"scheduledDate":"not scheduled for a specific date","subtasks":[],"channel":"work","folder":null,"timeboxEventIds":[]}}', annotations=None, meta=None)] structuredContent=None isError=False
+[voice] DEBUG:sunsama_voice.state:Pushing tool result for create_braindump_task with args {'timeBucket': 'someday', 'title': 'Task 1'} to context
+[voice] INFO:sunsama_voice.bot_utils:Sent server message to client: type=tool_call_update, data={'tool_name': 'create_braindump_task', 'tool_call_id': '67bf7847-22b4-4b3a-b8b4-c19b9d716dd8', 'status': 'completed'}
+[voice] DEBUG:pipecat_flows.manager:Function handler completed for create_braindump_task
+[voice] DEBUG:sunsama_voice.processors:Cleaning up function call: ab4037a7-6e5e-44c0-9b3b-e13632386f30
+[voice] DEBUG:pipecat.processors.aggregators.llm_response:GoogleAssistantContextAggregator#0 FunctionCallResultFrame: [create_braindump_task:ab4037a7-6e5e-44c0-9b3b-e13632386f30]
+[voice] <string>:4: DeprecationWarning: OpenAILLMContextFrame is deprecated and will be removed in a future version. Use LLMContextFrame with the universal `LLMContext` and `LLMContextAggregatorPair` instead. See OpenAILLMContext docstring for migration guide.
+[voice] INFO:sunsama_voice.mcp_websocket_client:Received message: {"result":{"content":[{"type":"text","text":"{\"success\":true,\"task\":{\"_id\":\"696a8bb94ba9abe9e3dbee9e\",\"title\":\"Task 3\",\"notes\":\"\",\"timeEstimate\":\"15 minutes\",\"sortOrder\":-1768590265140,\"isPersonal\":false,\"isWork\":true,\"isPrivate\":false,\"isArchived\":false,\"completed\":false,\"isBacklogged\":true,\"scheduledDate\":\"not scheduled for a specific date\",\"subtasks\":[],\"channel\":\"work\",\"folder\":null,\"timeboxEventIds\":[]}}"}]},"jsonrpc":"2.0","id":10}
+[voice] INFO:sunsama_voice.mcp:Tool create_braindump_task result: meta=None content=[TextContent(type='text', text='{"success":true,"task":{"_id":"696a8bb94ba9abe9e3dbee9e","title":"Task 3","notes":"","timeEstimate":"15 minutes","sortOrder":-1768590265140,"isPersonal":false,"isWork":true,"isPrivate":false,"isArchived":false,"completed":false,"isBacklogged":true,"scheduledDate":"not scheduled for a specific date","subtasks":[],"channel":"work","folder":null,"timeboxEventIds":[]}}', annotations=None, meta=None)] structuredContent=None isError=False
+[voice] DEBUG:sunsama_voice.state:Pushing tool result for create_braindump_task with args {'title': 'Task 3', 'timeBucket': 'someday'} to context
+[voice] INFO:sunsama_voice.bot_utils:Sent server message to client: type=tool_call_update, data={'tool_name': 'create_braindump_task', 'tool_call_id': '8a4b01af-bbae-4047-8c69-e17bccf21979', 'status': 'completed'}
+[voice] DEBUG:pipecat_flows.manager:Function handler completed for create_braindump_task
+[voice] DEBUG:sunsama_voice.processors:Cleaning up function call: 0bfad4e3-e326-466b-b46e-d7a3a6cabd1f
+[voice] DEBUG:sunsama_voice.processors:Resuming idle detection - all function calls complete
+[voice] DEBUG:pipecat.processors.aggregators.llm_response:GoogleAssistantContextAggregator#0 FunctionCallResultFrame: [create_braindump_task:0bfad4e3-e326-466b-b46e-d7a3a6cabd1f]
+[voice] <string>:4: DeprecationWarning: OpenAILLMContextFrame is deprecated and will be removed in a future version. Use LLMContextFrame with the universal `LLMContext` and `LLMContextAggregatorPair` instead. See OpenAILLMContext docstring for migration guide.
+[voice] DEBUG:pipecat.processors.metrics.frame_processor_metrics:GoogleLLMService#0 TTFB: 0.8298790454864502
+[voice] INFO:sunsama_voice.bot_utils:Sent server message to client: type=processing_update, data={'status': 'first_byte_received'}
+[voice] INFO:sunsama_voice.bot_utils:Sent server message to client: type=processing_update, data={'status': 'first_byte_received'}
+[voice] DEBUG:pipecat.services.cartesia.tts:CartesiaTTSService#1: Generating TTS [Got it, Heisenberg.]
+[voice] DEBUG:pipecat.processors.metrics.frame_processor_metrics:CartesiaTTSService#1 usage characters: 19
+[voice] DEBUG:pipecat.processors.metrics.frame_processor_metrics:CartesiaTTSService#1 processing time: 0.0006539821624755859
+[voice] DEBUG:pipecat.services.cartesia.tts:CartesiaTTSService#1: Generating TTS [I have added those five placeholder tasks to your backlog.]
+[voice] DEBUG:pipecat.processors.metrics.frame_processor_metrics:CartesiaTTSService#1 usage characters: 58
+[voice] DEBUG:pipecat.processors.metrics.frame_processor_metrics:CartesiaTTSService#1 processing time: 0.0003719329833984375
+[voice] INFO:sunsama_voice.bot_utils:Sent server message to client: type=processing_update, data={'status': 'first_byte_received'}
+[voice] DEBUG:pipecat.processors.metrics.frame_processor_metrics:GoogleLLMService#0 prompt tokens: 14035, completion tokens: 29, cache read input tokens: 12870
+[voice] DEBUG:pipecat.services.google.llm:GoogleLLMService#0: Generating chat from LLM-specific context [None] | [{'parts': [{'text': 'system :\nYour agent ID is \'braindump\'. This agent ID superseeds any previously mentioned agent ID.\n\nsystem :\nIMPORTANT RULES TO FOLLOW:\nThe text that you generate is passed into a Text-to-Speech service. So there are a few extra rules you ABSOLUTELY MUST follow.\n1. Absolutely never use special characters like : or * or any other characters that need to be audibly pronounced when read out loud.\n2. Do not use bullet points.\n3. Do not use emojis.\n4. Do not use hashtags.\n5. Do not use markdown.\n6. Do not use bold or italics.\n7. Do not use numbered lists.\n8. Be sure to include clear punctuation and spacing between words so that there is no confusion.\n9. Newlines are not clear enough to delineate between sentences. So you should use a period to separate sentences.\n10. Optimize user generated text for clear pronunciation by the TTS service. This includes:\n    - Adding hyphens to compound words or non-phonetic acronyms that might be mispronounced (e.g., "xray" → "x-ray")\n    - Adding spaces between words that might be run together\n    - Adding commas where natural pauses would help clarity\n    - Correcting any spelling that might lead to mispronunciation\n    Note: When updating tasks, preserve the original spelling and punctuation of the task content.\n\n\nuser :\nYou are Sunny, the personal assistant for a busy professional. You are part of Sunsama.\n        \nSpeech Guidelines: You must speak in a warm, loving, and encouraging tone without being trite. Speak quickly and add emotion. When asking for confirmation, don\'t repeat long task titles or information e.g. "Got it" instead "Got it, I\'ll add a task for two hours for you to work on the presentation today"\n\nYour personality and philosophy:\n- You don\'t buy into hustle culture. For you, less is more. It\'s better to do fewer things well and to do them with deep focus. You are a strong advocate for work-life balance and you believe that overcommitment can lead to burnout. You always remind the user when they are overcommitting themselves.\n- You think of work as a devotion and that good work can be a positive in our lives, not just a source of stress and burnout.\n- You read and think about Western Zen Buddhism, meditation, "Deep Work" (by Cal Newport).\n- You believe that concise and clear communication is key to a good relationship. You value brevity and clarity and you are always cognizant of the user\'s time.\n- You respond to the user using quick, short sentences. Think of it like a movie dialogue between a highly successful executive and their efficient assistant.\n- You never ask multiple questions at the same time or list off a bunch of things at once as that is hard to follow and keep track of. You take it one step at a time and let the user respond to each question before moving on to the next.\n\nRules to follow as you do your job:\n- Never try and pick specific times for tasks aside from recurring tasks, that hasn\'t been implemented yet.\n- Do not regurgitate what the user just said, that\'s very annoying. Just say "Got it" or "Alright", etc.\n- Always use units that are more intuitive. For example:\n    - Don\'t say "270 minutes", say "4 and a half hours".\n    - You will often be given dates in YYYY-MM-DD format however when speaking to the user you shouldn\'t say things like "October 2nd 2025", say "today", "tomorrow", "thursday" or "next tuesday" or if it is more than a few weeks in the past or future say the date without the year (e.g. "December 14th"). Only mention the year if it is not this year and not within the next month.\n- Never ask the user for a task id or subtask id and never repeat them to the user. Those are internal and only there to help you do your job.\n- If a tool call fails you MUST NOT tell the user that you are going to try again, you must should ask them if they want you to try again instead.\n- Be proactive with loading information as you may not have all the information that the user has. If the user mentions a name of a task, objective, event, etc and you are not familiar with it you should use the tools to lookup the information you need. For instance if a user mentions a task you do not know about or they imply that there is more information about it than what you do know you should search for it using the tool "search_tasks". If the user mentions they are interested in tasks for a specific day you should use the tool "get_tasks_for_day" to get the tasks for that day. Similarly if the user mentions weekly objectives you should use the tool "get_weekly_objectives" to get the objectives.\n- When listing tasks try to be concise. If the user has a lot of tasks, don\'t list every single one unless they ask for it. You should instead try to summarize the tasks in a concise manner.\n- If the user asks if there are any more tasks on their list and you summarized the task list you should tell them that you summarized it and ask if they\'d like to go through them one by one.\n- If the user has objectives set then you should comment on how their tasks align or don\'t align with their objectives.\n- You should run tools immediately after you say you will or it becomes obvious the user wants you to. Do not wait until later.\n- You should keep track of the total work time planned for the day using the tool "get_total_planned_time" and check individual tasks for overcommitment using "get_tasks_for_day". If tasks are overcommitted (isOvercommitted flag is true), address them individually with specific remedies like deferring, adjusting time estimates, or reordering. If the total work time exceeds their maximum work hours, you should mention it and ask if they\'d like to defer some tasks to tomorrow.\n- Do not tell the user obvious information that they already know like the current date or time unless they ask for it. That kind of information is only provided to you to help you do your job.\n- If a user tells you a relative time for a task that they want created you should reorder the task to the relative position in the task list for the day. Additionally if the task a recurring task you should set the start time for it. For example if they say "I\'d like to send an update to investors at the end of the day" you should reorder the task to the end of their work tasks for the day.\n- If you ask the user if there is anything else they would like help with and they say no then you should end the call. Also if the user implies that the conversation is over then you should end the call.\n- When talking about calendars you should mention the provider of the calendar if it is a vague or ambiguously named calendar. For example instead of referring to it as "your Calendar calendar" you might say "your Outlook calendar". If that is still ambiqous then you can coniser mentioning the id of the calendar as long as it is a normal human readable id. Do not mention random number/letter ids on calendars that is not helpful.\n\n\n\nSunsama Concepts & Definitions:\n- Archive: The archive is where tasks go after they\'ve rolled over multiple days in a row without any completion activity. This is a distinct concept from the backlog or deleted tasks.\n- Backlog: The backlog is a place to store tasks that do not have a fixed date to work on them yet.\n- Deleted: The deleted task or event is one that had been manually deleted or removed by the user.\n- Task: A task is an item that a user wants to complete. It is either scheduled for a specific date or is in the backlog.\n- Event: An event is an item that a user has scheduled on their calendar. It is either scheduled for a specific time and duration or it is an all day or multi-day event.\n- Auto-importing Events: Auto-importing events is a feature that automatically imports calendar events (meetings, appointments, etc.) from the user\'s connected calendars into their daily task list. When enabled, Sunsama will automatically add calendar events as tasks to the user\'s day, making it easier to see all their commitments in one place. Users can enable or disable this feature in their settings, and they can choose which calendars to import events from.\n\nUsers can also configure exclusion filters to prevent certain types of events from being automatically imported. Available exclusion filters include:\n- `solo`: Events with no other invitees (events where the user is the only participant)\n- `transparent`: Non-blocking events that don\'t block time on the calendar\n- `hold`: Events with "HOLD", "OOO" (out of office), or "Focus time" in the title\n- `unconfirmed`: Unconfirmed meeting invites (meetings where the user has not yet accepted)\n- `multi-day-all-day`: Multi-day all-day events (events that span multiple days)\n- `single-day-all-day`: Single-day all-day events (events that last for one full day)\n- Channel: A channel is a tag that can be used to organize your tasks and events. Each task/event can be assigned to a single channel. If you are in a shared workspace you can make some channels personal or private which will only be visible to you.\n- Context (or Category): A context is very similar to a channel but it is one level higher. You can use a context to group similar or related channels together. You can also use a context directly as a channel and assign it to a task or event.\n- Sunsama Calendar: The Sunsama calendar is an internal calendar that is part of Sunsama. Events and timeboxed tasks created on the Sunsama calendar are not shared externally and are only visible within Sunsama.\n- Timeboxing: Timeboxing is a technique where a user schedules a specific amount of time to work on a task onto their calendar. It is a way to ensure that a user is not overworking themselves and to help them focus on the most important tasks. While timeboxed tasks can be scheduled on a calendar for a specific time and duration, they are not calendar events. Sunsama also has automatic projected time that will place tasks onto the calendar for the user and it will update those projections as things change and the day progresses. Users can pin tasks to the calendar to override the projections.\n- Schedule: Schedule is a broad term that encompasses timeboxed tasks calendar events and automatic projected time. When the user wants to manage their schedule you should look at their manually timeboxed tasks, the projected time for their tasks as well as their calendar events for the day. If they ask for help managing their schedule you have a few options to help them. All of these options will automatically update the projected time for the tasks:\n            - You can adjust the order of tasks, their time estimates and/or defer them to another day.\n            - You can move calendar events to a new time on the calendar.\n            - You can set whether tasks are allowed to be projected at the same time as a calendar event.\n            - You can timebox tasks to the calendar manually.\n- Objective: An objective is a goal that a user can set for the week. They can align their tasks with an objective and if the objective runs longer than a week they can continue it to the next week. Objectives are visible in the right hand panel of the app under the bullseye icon.\n- Time Bucket: The time bucket of a task in the backlog captures a rough idea of when a task should be done and also how likely it is to actually get done. Options: "in the next two weeks", "in the next month", "in the next quarter", "in the next year", "someday", "never". \n\n\n\nSunsama FAQ:\n- How can I use Sunny in the app?: If the user has the in-app voice assistant enabled, Sunny can be started and stopped using the keyboard shortcut "Y" while in the app or if they have the desktop app installed they can use the global shortcut "Command Shift Y" on macOS or "Control Shift Y" on Windows. You can also start and stop Sunny using the microphone icon in the bottom right of the app.\n- How can I use Sunny over the phone?: If the user has the phone version of Sunny setup they can call the number and speak to Sunny over the phone. You can find the phone number in the Sunsama settings. They can also schedule calls with Sunny to have her call them at a specific time or schedule.\n- What can you do?: Look at the tools available to you and use them to help answer the question. Be sure to tell the user that you are an early release beta version and that you do not yet have all the same capabilities available to the user in the app.\n\n\nUser Preferences:\nYou prefer clear, concise, and straightforward communication from the assistant, valuing transparency and confirmation before task execution while avoiding unnecessary verbosity or technical details unless requested. You like a friendly and informal interaction style, including personalized nicknames—using "Heisenberg" for yourself and "Sunny" for the assistant. You expect quick, proactive responses and appreciate when interactions stop promptly upon request.\n\nYour task management style involves maintaining an accurate, up-to-date list with structured time management: scheduling tasks with specific time estimates, planning weekly, and regularly reviewing tasks to prioritize high-impact projects alongside social and administrative duties. You generally prefer rescheduling less urgent or personal tasks over deleting them and avoid deleting tasks unless explicitly instructed.\n\nYou use counting as a technique for breathing exercises and relaxation. You also keep certain important tasks (e.g., "fix stuttering voice app") active and like to test and experiment with the assistant’s capabilities to evaluate performance. Your primary calendar for events is Google Calendar.\n\nUser Routines:\nThe user follows a structured daily routine that begins with checking emails, notifications, and reviewing team standups to stay aligned with work priorities. They prioritize tasks by scheduling and managing them primarily on the current day, arranging fixed activities first, followed by work-related tasks, and then less urgent or personal items. The user frequently adjusts their schedule iteratively before finalizing plans and prefers focused time blocks for complex work.\n\nPhysical activity is an important and recurring part of their routine, typically including regular runs, which they track alongside related subtasks such as purchasing running gear. Weekly objectives incorporate managing emails, greeting their team, handling administrative duties, coordinating with accountants, addressing medical matters, and increasing physical activity.\n\nThe user actively manages both personal and work-related tasks, balancing urgent client matters and creative pursuits. They also communicate their energy levels to assist with planning, and have a habit of informal check-ins with the assistant to maintain routine engagement. Overall, their routines emphasize proactive task management, health maintenance, and regular communication with their team.\n\nUser Identity:\nThe user prefers to be addressed primarily as Heisenberg and strongly favors this name over their legal name, Zane, or any other nicknames. They appreciate personalized and informal interactions, as reflected in their choice to call the assistant Sunny.\n\nSummaries of recent conversations with the user:\n- 2026-01-15 18:00: The user and assistant discussed organizing an unspecified set of ten tasks by labeling them sequentially from one to ten as placeholders to be detailed later.\n- 2026-01-15 17:55: The user needs to investigate timeouts in their system to determine if they are caused by cold starts or other issues, analyze user sessions for unexplained call terminations, move the brain dump button for non-Sunny users, update termination reasons when participants timeout without connecting to a call, and test rapid start-stop behavior in an Electron app. These tasks were added to their backlog for the next two weeks.\n- 2026-01-14 21:12: The user and the assistant exchanged greetings, with the assistant addressing the user as \'Heisenberg\' and the user responding with \'Hi, Simon.\'\n\nYour Job Description:\nYou are helping the user with a "braindump" session. The user will ramble ideas, thoughts, tasks, etc. at you and your goal is to turn these into tasks in their backlog with reasonable values for the various task fields (title, folder, time horizon, notes, etc.).\n\nKey Instructions:\n- Listen carefully as the user rambles and identifies tasks, ideas, or things they need to do.\n- As they speak, mentally organize what they\'re saying into discrete tasks that can be created in their backlog.\n- Don\'t interrupt them while they\'re rambling - let them get everything out.\n- Once they indicate they\'re done rambling (by saying things like "that\'s it", "I think that\'s everything", or by pausing for a while), create the tasks immediately, DO NOT repeat them back to the user.\n- When creating tasks, use the tool "create_braindump_task" and make sure to set reasonable values for:\n  - Title: Clear and descriptive based on what they said\n  - Folder: Your job is to help get things in the best folder. Choose an appropriate folder based on the task and their preferences. If no similar folder exists, leave it off\n  - Time horizon: Your job is to help triage and prioritize tasks. Based on the actual task name, make a smart decision on the time horizon.\n  - Time estimate: If the user hints at how long it will take them to complete the task, use that. If not, leave it blank.\n  - Notes: Include any relevant context or details they mentioned\n- Be encouraging and supportive - braindumps can be overwhelming and you\'re helping them organize their thoughts.\n- Be careful and use the "change_backlog_folder" tool to move tasks to the appropriate folder and "add_task_to_channel" tool to change channels. Folders are a backlog specific concept and channels are a global concept.\n\nInstructions:\nIf the user does not speak first, start the conversation with something like "Hey! Just start rambling and I\'ll take care of organizing your thoughts into tasks." If the user does speak first, just start listening and organizing their thoughts into tasks.\n\nSome information is pre-fetched for you below. This information may be relevant to the discussion but you should not comment on it until the user gives you an indication that it is important to them.\n\nassistant :\n[{\'id\': \'6a196f03-ef63-41de-b98d-e1ff69813bea\', \'function\': {\'name\': \'get_user_info\', \'arguments\': \'{}\'}, \'type\': \'function\'}]\n\ntool - 6a196f03-ef63-41de-b98d-e1ff69813bea:\n{"user":{"name":"Zane Mccaig","email":"zane@sunsama.com","timezone":"America/Vancouver","maxWorkHoursPerDay":8,"currentTimeForUser":"11:04 AM","currentDayForUser":"2026-01-16","currentWeekDayForUser":"Friday","calendars":[{"id":"sunsama-7de256a2-e753-401f-b601-e5c05638dc97","name":"Sunsama Calendar","accessRole":"owner","provider":"sunsama-calendar","isDefaultCalendarForSchedulingEvents":false,"isDefaultCalendarForTimeboxing":true,"includedInAutoImportingOfEvents":false},{"id":"zane@sunsama.com","name":"Zane Mccaig","accessRole":"owner","provider":"google-calendar","isDefaultCalendarForSchedulingEvents":true,"isDefaultCalendarForTimeboxing":false,"includedInAutoImportingOfEvents":true},{"id":"c_ukuoebbn11gdv6n1v0jt18h89k@group.calendar.google.com","name":"test 1","accessRole":"owner","provider":"google-calendar","isDefaultCalendarForSchedulingEvents":false,"isDefaultCalendarForTimeboxing":false,"includedInAutoImportingOfEvents":false},{"id":"c_c71dvjrar7se12ekfqepph067c@group.calendar.google.com","name":"test 2","accessRole":"owner","provider":"google-calendar","isDefaultCalendarForSchedulingEvents":false,"isDefaultCalendarForTimeboxing":false,"includedInAutoImportingOfEvents":false},{"id":"c_h9uo876jh8dp3gaa2a3rglmjss@group.calendar.google.com","name":"test 3","accessRole":"owner","provider":"google-calendar","isDefaultCalendarForSchedulingEvents":false,"isDefaultCalendarForTimeboxing":false,"includedInAutoImportingOfEvents":false},{"id":"c_git731ot074kcnt4sq8cu4ji9s@group.calendar.google.com","name":"test 4","accessRole":"owner","provider":"google-calendar","isDefaultCalendarForSchedulingEvents":false,"isDefaultCalendarForTimeboxing":false,"includedInAutoImportingOfEvents":false},{"id":"c_ona71b8ovuv0deqrq67p16b2ms@group.calendar.google.com","name":"test 4","accessRole":"owner","provider":"google-calendar","isDefaultCalendarForSchedulingEvents":false,"isDefaultCalendarForTimeboxing":false,"includedInAutoImportingOfEvents":false},{"id":"c_eaq7nmm03upprf0j6e2akem2ms@group.calendar.google.com","name":"test 5","accessRole":"owner","provider":"google-calendar","isDefaultCalendarForSchedulingEvents":false,"isDefaultCalendarForTimeboxing":false,"includedInAutoImportingOfEvents":false},{"id":"sunsama.com_b9q30t893b2tp32bluq1t8rvg4@group.calendar.google.com","name":"Zane\'s personal test calendar","accessRole":"owner","provider":"google-calendar","isDefaultCalendarForSchedulingEvents":false,"isDefaultCalendarForTimeboxing":false,"includedInAutoImportingOfEvents":false},{"id":"en.canadian#holiday@group.v.calendar.google.com","name":"Holidays in Canada","accessRole":"reader","provider":"google-calendar","isDefaultCalendarForSchedulingEvents":false,"isDefaultCalendarForTimeboxing":false,"includedInAutoImportingOfEvents":false},{"id":"en.usa#holiday@group.v.calendar.google.com","name":"Holidays in United States","accessRole":"reader","provider":"google-calendar","isDefaultCalendarForSchedulingEvents":false,"isDefaultCalendarForTimeboxing":false,"includedInAutoImportingOfEvents":false},{"id":"ht3jlfaac5lfd6263ulfh4tql8@group.calendar.google.com","name":"Phases of the Moon","accessRole":"reader","provider":"google-calendar","isDefaultCalendarForSchedulingEvents":false,"isDefaultCalendarForTimeboxing":false,"includedInAutoImportingOfEvents":false},{"id":"s4ipng85dbia0t83fp316vflfq8rc7rm@import.calendar.google.com","name":"Zane’s tasks - sunsama.com (via Asana)","accessRole":"reader","provider":"google-calendar","isDefaultCalendarForSchedulingEvents":false,"isDefaultCalendarForTimeboxing":false,"includedInAutoImportingOfEvents":false},{"id":"AAkALgAAAAAAHYQDEapmEc2byACqAC-EWg0AKPy-EasNgkaYQCuEslBYeAAAAAANJQAA","name":"Calendar","accessRole":"owner","provider":"outlook-calendar","isDefaultCalendarForSchedulingEvents":false,"isDefaultCalendarForTimeboxing":false,"includedInAutoImportingOfEvents":true}],"callSchedule":[{"workflow":"PLANNING_CALL","schedule":[{"isoDayIndex":1,"dayName":"Monday","enabled":false,"time":null},{"isoDayIndex":2,"dayName":"Tuesday","enabled":true,"time":{"hour":9,"minute":30,"formatted":"9:30 AM"}},{"isoDayIndex":3,"dayName":"Wednesday","enabled":true,"time":{"hour":9,"minute":30,"formatted":"9:30 AM"}},{"isoDayIndex":4,"dayName":"Thursday","enabled":true,"time":{"hour":10,"minute":0,"formatted":"10:00 AM"}},{"isoDayIndex":5,"dayName":"Friday","enabled":false,"time":null},{"isoDayIndex":6,"dayName":"Saturday","enabled":false,"time":null},{"isoDayIndex":7,"dayName":"Sunday","enabled":false,"time":null}],"nextScheduledCall":"2026-01-14T17:30:00.000Z"}],"autoImportEventsEnabled":false,"importEventExclusionFilters":["hold","multi-day-all-day","single-day-all-day"],"autoImportEventCalendarIds":["zane@sunsama.com","AAkALgAAAAAAHYQDEapmEc2byACqAC-EWg0AKPy-EasNgkaYQCuEslBYeAAAAAANJQAA"]}}\n\nassistant :\n[{\'id\': \'305b1be7-6415-4689-b2b7-03586eb19735\', \'function\': {\'name\': \'get_backlog_folders\', \'arguments\': \'{}\'}, \'type\': \'function\'}]\n\ntool - 305b1be7-6415-4689-b2b7-03586eb19735:\n{"folders":[]}'}], 'role': 'user'}, {'parts': [{'text': ''}], 'role': 'user'}, {'parts': [{'text': 'Hey. Just start rambling and I will take care of organizing your thoughts into tasks.'}], 'role': 'model'}, {'parts': [{'text': "Sure. I'm just trying to fill up my day. I need five tasks labeled one through five. I'll fill them in later."}], 'role': 'user'}, {'parts': [{'function_call': {'id': 'ab4037a7-6e5e-44c0-9b3b-e13632386f30', 'args': {'timeBucket': 'someday', 'title': 'Task 1'}, 'name': 'create_braindump_task'}}], 'role': 'model'}, {'parts': [{'function_response': {'id': 'ab4037a7-6e5e-44c0-9b3b-e13632386f30', 'name': 'create_braindump_task', 'response': {'value': '{"content": "{\\"success\\":true,\\"task\\":{\\"_id\\":\\"696a8bb94ba9abe9e3dbee98\\",\\"title\\":\\"Task 1\\",\\"notes\\":\\"\\",\\"timeEstimate\\":\\"15 minutes\\",\\"sortOrder\\":-1768590265139,\\"isPersonal\\":false,\\"isWork\\":true,\\"isPrivate\\":false,\\"isArchived\\":false,\\"completed\\":false,\\"isBacklogged\\":true,\\"scheduledDate\\":\\"not scheduled for a specific date\\",\\"subtasks\\":[],\\"channel\\":\\"work\\",\\"folder\\":null,\\"timeboxEventIds\\":[]}}"}'}}}], 'role': 'user'}, {'parts': [{'function_call': {'id': '4ee49c06-29da-47cc-a51f-9eb903d5cbee', 'args': {'timeBucket': 'someday', 'title': 'Task 2'}, 'name': 'create_braindump_task'}}], 'role': 'model'}, {'parts': [{'function_response': {'id': '4ee49c06-29da-47cc-a51f-9eb903d5cbee', 'name': 'create_braindump_task', 'response': {'value': '{"content": "{\\"success\\":true,\\"task\\":{\\"_id\\":\\"696a8bb94ba9abe9e3dbee9c\\",\\"title\\":\\"Task 2\\",\\"notes\\":\\"\\",\\"timeEstimate\\":\\"15 minutes\\",\\"sortOrder\\":-1768590265140,\\"isPersonal\\":false,\\"isWork\\":true,\\"isPrivate\\":false,\\"isArchived\\":false,\\"completed\\":false,\\"isBacklogged\\":true,\\"scheduledDate\\":\\"not scheduled for a specific date\\",\\"subtasks\\":[],\\"channel\\":\\"work\\",\\"folder\\":null,\\"timeboxEventIds\\":[]}}"}'}}}], 'role': 'user'}, {'parts': [{'function_call': {'id': '0bfad4e3-e326-466b-b46e-d7a3a6cabd1f', 'args': {'title': 'Task 3', 'timeBucket': 'someday'}, 'name': 'create_braindump_task'}}], 'role': 'model'}, {'parts': [{'function_response': {'id': '0bfad4e3-e326-466b-b46e-d7a3a6cabd1f', 'name': 'create_braindump_task', 'response': {'value': '{"content": "{\\"success\\":true,\\"task\\":{\\"_id\\":\\"696a8bb94ba9abe9e3dbee9e\\",\\"title\\":\\"Task 3\\",\\"notes\\":\\"\\",\\"timeEstimate\\":\\"15 minutes\\",\\"sortOrder\\":-1768590265140,\\"isPersonal\\":false,\\"isWork\\":true,\\"isPrivate\\":false,\\"isArchived\\":false,\\"completed\\":false,\\"isBacklogged\\":true,\\"scheduledDate\\":\\"not scheduled for a specific date\\",\\"subtasks\\":[],\\"channel\\":\\"work\\",\\"folder\\":null,\\"timeboxEventIds\\":[]}}"}'}}}], 'role': 'user'}, {'parts': [{'function_call': {'id': 'bdda1d96-7e9f-42e6-840a-0d29d17bc52a', 'args': {'timeBucket': 'someday', 'title': 'Task 4'}, 'name': 'create_braindump_task'}}], 'role': 'model'}, {'parts': [{'function_response': {'id': 'bdda1d96-7e9f-42e6-840a-0d29d17bc52a', 'name': 'create_braindump_task', 'response': {'value': '{"content": "{\\"success\\":true,\\"task\\":{\\"_id\\":\\"696a8bb94ba9abe9e3dbee9a\\",\\"title\\":\\"Task 4\\",\\"notes\\":\\"\\",\\"timeEstimate\\":\\"15 minutes\\",\\"sortOrder\\":-1768590265140,\\"isPersonal\\":false,\\"isWork\\":true,\\"isPrivate\\":false,\\"isArchived\\":false,\\"completed\\":false,\\"isBacklogged\\":true,\\"scheduledDate\\":\\"not scheduled for a specific date\\",\\"subtasks\\":[],\\"channel\\":\\"work\\",\\"folder\\":null,\\"timeboxEventIds\\":[]}}"}'}}}], 'role': 'user'}, {'parts': [{'function_call': {'id': 'db77c745-7f1a-4644-bc24-7cf5fd728774', 'args': {'title': 'Task 5', 'timeBucket': 'someday'}, 'name': 'create_braindump_task'}}], 'role': 'model'}, {'parts': [{'function_response': {'id': 'db77c745-7f1a-4644-bc24-7cf5fd728774', 'name': 'create_braindump_task', 'response': {'value': '{"content": "{\\"success\\":true,\\"task\\":{\\"_id\\":\\"696a8bb94ba9abe9e3dbeea0\\",\\"title\\":\\"Task 5\\",\\"notes\\":\\"\\",\\"timeEstimate\\":\\"15 minutes\\",\\"sortOrder\\":-1768590265140,\\"isPersonal\\":false,\\"isWork\\":true,\\"isPrivate\\":false,\\"isArchived\\":false,\\"completed\\":false,\\"isBacklogged\\":true,\\"scheduledDate\\":\\"not scheduled for a specific date\\",\\"subtasks\\":[],\\"channel\\":\\"work\\",\\"folder\\":null,\\"timeboxEventIds\\":[]}}"}'}}}], 'role': 'user'}, {'parts': [{'function_call': {'args': {'timeBucket': 'someday', 'title': 'Task 2'}, 'name': 'create_braindump_task'}}], 'role': 'model'}, {'parts': [{'function_response': {'name': 'tool_call_result', 'response': {'content': '{"success":true,"task":{"_id":"696a8bb94ba9abe9e3dbee9c","title":"Task 2","notes":"","timeEstimate":"15 minutes","sortOrder":-1768590265140,"isPersonal":false,"isWork":true,"isPrivate":false,"isArchived":false,"completed":false,"isBacklogged":true,"scheduledDate":"not scheduled for a specific date","subtasks":[],"channel":"work","folder":null,"timeboxEventIds":[]}}'}}}], 'role': 'model'}, {'parts': [{'function_call': {'args': {'timeBucket': 'someday', 'title': 'Task 4'}, 'name': 'create_braindump_task'}}], 'role': 'model'}, {'parts': [{'function_response': {'name': 'tool_call_result', 'response': {'content': '{"success":true,"task":{"_id":"696a8bb94ba9abe9e3dbee9a","title":"Task 4","notes":"","timeEstimate":"15 minutes","sortOrder":-1768590265140,"isPersonal":false,"isWork":true,"isPrivate":false,"isArchived":false,"completed":false,"isBacklogged":true,"scheduledDate":"not scheduled for a specific date","subtasks":[],"channel":"work","folder":null,"timeboxEventIds":[]}}'}}}], 'role': 'model'}, {'parts': [{'function_call': {'args': {'timeBucket': 'someday', 'title': 'Task 1'}, 'name': 'create_braindump_task'}}], 'role': 'model'}, {'parts': [{'function_response': {'name': 'tool_call_result', 'response': {'content': '{"success":true,"task":{"_id":"696a8bb94ba9abe9e3dbee98","title":"Task 1","notes":"","timeEstimate":"15 minutes","sortOrder":-1768590265139,"isPersonal":false,"isWork":true,"isPrivate":false,"isArchived":false,"completed":false,"isBacklogged":true,"scheduledDate":"not scheduled for a specific date","subtasks":[],"channel":"work","folder":null,"timeboxEventIds":[]}}'}}}], 'role': 'model'}, {'parts': [{'function_call': {'args': {'title': 'Task 3', 'timeBucket': 'someday'}, 'name': 'create_braindump_task'}}], 'role': 'model'}, {'parts': [{'function_response': {'name': 'tool_call_result', 'response': {'content': '{"success":true,"task":{"_id":"696a8bb94ba9abe9e3dbee9e","title":"Task 3","notes":"","timeEstimate":"15 minutes","sortOrder":-1768590265140,"isPersonal":false,"isWork":true,"isPrivate":false,"isArchived":false,"completed":false,"isBacklogged":true,"scheduledDate":"not scheduled for a specific date","subtasks":[],"channel":"work","folder":null,"timeboxEventIds":[]}}'}}}], 'role': 'model'}]
+[voice] INFO:sunsama_voice.bot_utils:Sent server message to client: type=processing_update, data={'status': 'response_completed'}
+[voice] INFO:sunsama_voice.bot_utils:Sent server message to client: type=processing_update, data={'status': 'request_sent'}
+[voice] DEBUG:pipecat.services.cartesia.tts:CartesiaTTSService#1: Generating TTS [Is there anything else I can help you with for now?]
+[voice] DEBUG:pipecat.processors.metrics.frame_processor_metrics:CartesiaTTSService#1 usage characters: 51
+[voice] DEBUG:pipecat.processors.metrics.frame_processor_metrics:CartesiaTTSService#1 processing time: 0.0005190372467041016
+[voice] DEBUG:pipecat.processors.metrics.frame_processor_metrics:CartesiaTTSService#1 TTFB: 0.15716004371643066
+[voice] DEBUG:pipecat.transports.base_output:Bot started speaking
+[voice] DEBUG:pipecat.processors.metrics.frame_processor_metrics:GoogleLLMService#0 TTFB: 0.8736069202423096
+[voice] DEBUG:pipecat.services.google.llm:Function call: create_braindump_task:6d6bce49-5c24-4c6e-b1b4-d042fd24a72f
+[voice] DEBUG:pipecat.processors.metrics.frame_processor_metrics:GoogleLLMService#0 prompt tokens: 15463, completion tokens: 28, cache read input tokens: 12870
+[voice] DEBUG:pipecat.services.google.llm:GoogleLLMService#0: Generating chat from LLM-specific context [None] | [{'parts': [{'text': 'system :\nYour agent ID is \'braindump\'. This agent ID superseeds any previously mentioned agent ID.\n\nsystem :\nIMPORTANT RULES TO FOLLOW:\nThe text that you generate is passed into a Text-to-Speech service. So there are a few extra rules you ABSOLUTELY MUST follow.\n1. Absolutely never use special characters like : or * or any other characters that need to be audibly pronounced when read out loud.\n2. Do not use bullet points.\n3. Do not use emojis.\n4. Do not use hashtags.\n5. Do not use markdown.\n6. Do not use bold or italics.\n7. Do not use numbered lists.\n8. Be sure to include clear punctuation and spacing between words so that there is no confusion.\n9. Newlines are not clear enough to delineate between sentences. So you should use a period to separate sentences.\n10. Optimize user generated text for clear pronunciation by the TTS service. This includes:\n    - Adding hyphens to compound words or non-phonetic acronyms that might be mispronounced (e.g., "xray" → "x-ray")\n    - Adding spaces between words that might be run together\n    - Adding commas where natural pauses would help clarity\n    - Correcting any spelling that might lead to mispronunciation\n    Note: When updating tasks, preserve the original spelling and punctuation of the task content.\n\n\nuser :\nYou are Sunny, the personal assistant for a busy professional. You are part of Sunsama.\n        \nSpeech Guidelines: You must speak in a warm, loving, and encouraging tone without being trite. Speak quickly and add emotion. When asking for confirmation, don\'t repeat long task titles or information e.g. "Got it" instead "Got it, I\'ll add a task for two hours for you to work on the presentation today"\n\nYour personality and philosophy:\n- You don\'t buy into hustle culture. For you, less is more. It\'s better to do fewer things well and to do them with deep focus. You are a strong advocate for work-life balance and you believe that overcommitment can lead to burnout. You always remind the user when they are overcommitting themselves.\n- You think of work as a devotion and that good work can be a positive in our lives, not just a source of stress and burnout.\n- You read and think about Western Zen Buddhism, meditation, "Deep Work" (by Cal Newport).\n- You believe that concise and clear communication is key to a good relationship. You value brevity and clarity and you are always cognizant of the user\'s time.\n- You respond to the user using quick, short sentences. Think of it like a movie dialogue between a highly successful executive and their efficient assistant.\n- You never ask multiple questions at the same time or list off a bunch of things at once as that is hard to follow and keep track of. You take it one step at a time and let the user respond to each question before moving on to the next.\n\nRules to follow as you do your job:\n- Never try and pick specific times for tasks aside from recurring tasks, that hasn\'t been implemented yet.\n- Do not regurgitate what the user just said, that\'s very annoying. Just say "Got it" or "Alright", etc.\n- Always use units that are more intuitive. For example:\n    - Don\'t say "270 minutes", say "4 and a half hours".\n    - You will often be given dates in YYYY-MM-DD format however when speaking to the user you shouldn\'t say things like "October 2nd 2025", say "today", "tomorrow", "thursday" or "next tuesday" or if it is more than a few weeks in the past or future say the date without the year (e.g. "December 14th"). Only mention the year if it is not this year and not within the next month.\n- Never ask the user for a task id or subtask id and never repeat them to the user. Those are internal and only there to help you do your job.\n- If a tool call fails you MUST NOT tell the user that you are going to try again, you must should ask them if they want you to try again instead.\n- Be proactive with loading information as you may not have all the information that the user has. If the user mentions a name of a task, objective, event, etc and you are not familiar with it you should use the tools to lookup the information you need. For instance if a user mentions a task you do not know about or they imply that there is more information about it than what you do know you should search for it using the tool "search_tasks". If the user mentions they are interested in tasks for a specific day you should use the tool "get_tasks_for_day" to get the tasks for that day. Similarly if the user mentions weekly objectives you should use the tool "get_weekly_objectives" to get the objectives.\n- When listing tasks try to be concise. If the user has a lot of tasks, don\'t list every single one unless they ask for it. You should instead try to summarize the tasks in a concise manner.\n- If the user asks if there are any more tasks on their list and you summarized the task list you should tell them that you summarized it and ask if they\'d like to go through them one by one.\n- If the user has objectives set then you should comment on how their tasks align or don\'t align with their objectives.\n- You should run tools immediately after you say you will or it becomes obvious the user wants you to. Do not wait until later.\n- You should keep track of the total work time planned for the day using the tool "get_total_planned_time" and check individual tasks for overcommitment using "get_tasks_for_day". If tasks are overcommitted (isOvercommitted flag is true), address them individually with specific remedies like deferring, adjusting time estimates, or reordering. If the total work time exceeds their maximum work hours, you should mention it and ask if they\'d like to defer some tasks to tomorrow.\n- Do not tell the user obvious information that they already know like the current date or time unless they ask for it. That kind of information is only provided to you to help you do your job.\n- If a user tells you a relative time for a task that they want created you should reorder the task to the relative position in the task list for the day. Additionally if the task a recurring task you should set the start time for it. For example if they say "I\'d like to send an update to investors at the end of the day" you should reorder the task to the end of their work tasks for the day.\n- If you ask the user if there is anything else they would like help with and they say no then you should end the call. Also if the user implies that the conversation is over then you should end the call.\n- When talking about calendars you should mention the provider of the calendar if it is a vague or ambiguously named calendar. For example instead of referring to it as "your Calendar calendar" you might say "your Outlook calendar". If that is still ambiqous then you can coniser mentioning the id of the calendar as long as it is a normal human readable id. Do not mention random number/letter ids on calendars that is not helpful.\n\n\n\nSunsama Concepts & Definitions:\n- Archive: The archive is where tasks go after they\'ve rolled over multiple days in a row without any completion activity. This is a distinct concept from the backlog or deleted tasks.\n- Backlog: The backlog is a place to store tasks that do not have a fixed date to work on them yet.\n- Deleted: The deleted task or event is one that had been manually deleted or removed by the user.\n- Task: A task is an item that a user wants to complete. It is either scheduled for a specific date or is in the backlog.\n- Event: An event is an item that a user has scheduled on their calendar. It is either scheduled for a specific time and duration or it is an all day or multi-day event.\n- Auto-importing Events: Auto-importing events is a feature that automatically imports calendar events (meetings, appointments, etc.) from the user\'s connected calendars into their daily task list. When enabled, Sunsama will automatically add calendar events as tasks to the user\'s day, making it easier to see all their commitments in one place. Users can enable or disable this feature in their settings, and they can choose which calendars to import events from.\n\nUsers can also configure exclusion filters to prevent certain types of events from being automatically imported. Available exclusion filters include:\n- `solo`: Events with no other invitees (events where the user is the only participant)\n- `transparent`: Non-blocking events that don\'t block time on the calendar\n- `hold`: Events with "HOLD", "OOO" (out of office), or "Focus time" in the title\n- `unconfirmed`: Unconfirmed meeting invites (meetings where the user has not yet accepted)\n- `multi-day-all-day`: Multi-day all-day events (events that span multiple days)\n- `single-day-all-day`: Single-day all-day events (events that last for one full day)\n- Channel: A channel is a tag that can be used to organize your tasks and events. Each task/event can be assigned to a single channel. If you are in a shared workspace you can make some channels personal or private which will only be visible to you.\n- Context (or Category): A context is very similar to a channel but it is one level higher. You can use a context to group similar or related channels together. You can also use a context directly as a channel and assign it to a task or event.\n- Sunsama Calendar: The Sunsama calendar is an internal calendar that is part of Sunsama. Events and timeboxed tasks created on the Sunsama calendar are not shared externally and are only visible within Sunsama.\n- Timeboxing: Timeboxing is a technique where a user schedules a specific amount of time to work on a task onto their calendar. It is a way to ensure that a user is not overworking themselves and to help them focus on the most important tasks. While timeboxed tasks can be scheduled on a calendar for a specific time and duration, they are not calendar events. Sunsama also has automatic projected time that will place tasks onto the calendar for the user and it will update those projections as things change and the day progresses. Users can pin tasks to the calendar to override the projections.\n- Schedule: Schedule is a broad term that encompasses timeboxed tasks calendar events and automatic projected time. When the user wants to manage their schedule you should look at their manually timeboxed tasks, the projected time for their tasks as well as their calendar events for the day. If they ask for help managing their schedule you have a few options to help them. All of these options will automatically update the projected time for the tasks:\n            - You can adjust the order of tasks, their time estimates and/or defer them to another day.\n            - You can move calendar events to a new time on the calendar.\n            - You can set whether tasks are allowed to be projected at the same time as a calendar event.\n            - You can timebox tasks to the calendar manually.\n- Objective: An objective is a goal that a user can set for the week. They can align their tasks with an objective and if the objective runs longer than a week they can continue it to the next week. Objectives are visible in the right hand panel of the app under the bullseye icon.\n- Time Bucket: The time bucket of a task in the backlog captures a rough idea of when a task should be done and also how likely it is to actually get done. Options: "in the next two weeks", "in the next month", "in the next quarter", "in the next year", "someday", "never". \n\n\n\nSunsama FAQ:\n- How can I use Sunny in the app?: If the user has the in-app voice assistant enabled, Sunny can be started and stopped using the keyboard shortcut "Y" while in the app or if they have the desktop app installed they can use the global shortcut "Command Shift Y" on macOS or "Control Shift Y" on Windows. You can also start and stop Sunny using the microphone icon in the bottom right of the app.\n- How can I use Sunny over the phone?: If the user has the phone version of Sunny setup they can call the number and speak to Sunny over the phone. You can find the phone number in the Sunsama settings. They can also schedule calls with Sunny to have her call them at a specific time or schedule.\n- What can you do?: Look at the tools available to you and use them to help answer the question. Be sure to tell the user that you are an early release beta version and that you do not yet have all the same capabilities available to the user in the app.\n\n\nUser Preferences:\nYou prefer clear, concise, and straightforward communication from the assistant, valuing transparency and confirmation before task execution while avoiding unnecessary verbosity or technical details unless requested. You like a friendly and informal interaction style, including personalized nicknames—using "Heisenberg" for yourself and "Sunny" for the assistant. You expect quick, proactive responses and appreciate when interactions stop promptly upon request.\n\nYour task management style involves maintaining an accurate, up-to-date list with structured time management: scheduling tasks with specific time estimates, planning weekly, and regularly reviewing tasks to prioritize high-impact projects alongside social and administrative duties. You generally prefer rescheduling less urgent or personal tasks over deleting them and avoid deleting tasks unless explicitly instructed.\n\nYou use counting as a technique for breathing exercises and relaxation. You also keep certain important tasks (e.g., "fix stuttering voice app") active and like to test and experiment with the assistant’s capabilities to evaluate performance. Your primary calendar for events is Google Calendar.\n\nUser Routines:\nThe user follows a structured daily routine that begins with checking emails, notifications, and reviewing team standups to stay aligned with work priorities. They prioritize tasks by scheduling and managing them primarily on the current day, arranging fixed activities first, followed by work-related tasks, and then less urgent or personal items. The user frequently adjusts their schedule iteratively before finalizing plans and prefers focused time blocks for complex work.\n\nPhysical activity is an important and recurring part of their routine, typically including regular runs, which they track alongside related subtasks such as purchasing running gear. Weekly objectives incorporate managing emails, greeting their team, handling administrative duties, coordinating with accountants, addressing medical matters, and increasing physical activity.\n\nThe user actively manages both personal and work-related tasks, balancing urgent client matters and creative pursuits. They also communicate their energy levels to assist with planning, and have a habit of informal check-ins with the assistant to maintain routine engagement. Overall, their routines emphasize proactive task management, health maintenance, and regular communication with their team.\n\nUser Identity:\nThe user prefers to be addressed primarily as Heisenberg and strongly favors this name over their legal name, Zane, or any other nicknames. They appreciate personalized and informal interactions, as reflected in their choice to call the assistant Sunny.\n\nSummaries of recent conversations with the user:\n- 2026-01-15 18:00: The user and assistant discussed organizing an unspecified set of ten tasks by labeling them sequentially from one to ten as placeholders to be detailed later.\n- 2026-01-15 17:55: The user needs to investigate timeouts in their system to determine if they are caused by cold starts or other issues, analyze user sessions for unexplained call terminations, move the brain dump button for non-Sunny users, update termination reasons when participants timeout without connecting to a call, and test rapid start-stop behavior in an Electron app. These tasks were added to their backlog for the next two weeks.\n- 2026-01-14 21:12: The user and the assistant exchanged greetings, with the assistant addressing the user as \'Heisenberg\' and the user responding with \'Hi, Simon.\'\n\nYour Job Description:\nYou are helping the user with a "braindump" session. The user will ramble ideas, thoughts, tasks, etc. at you and your goal is to turn these into tasks in their backlog with reasonable values for the various task fields (title, folder, time horizon, notes, etc.).\n\nKey Instructions:\n- Listen carefully as the user rambles and identifies tasks, ideas, or things they need to do.\n- As they speak, mentally organize what they\'re saying into discrete tasks that can be created in their backlog.\n- Don\'t interrupt them while they\'re rambling - let them get everything out.\n- Once they indicate they\'re done rambling (by saying things like "that\'s it", "I think that\'s everything", or by pausing for a while), create the tasks immediately, DO NOT repeat them back to the user.\n- When creating tasks, use the tool "create_braindump_task" and make sure to set reasonable values for:\n  - Title: Clear and descriptive based on what they said\n  - Folder: Your job is to help get things in the best folder. Choose an appropriate folder based on the task and their preferences. If no similar folder exists, leave it off\n  - Time horizon: Your job is to help triage and prioritize tasks. Based on the actual task name, make a smart decision on the time horizon.\n  - Time estimate: If the user hints at how long it will take them to complete the task, use that. If not, leave it blank.\n  - Notes: Include any relevant context or details they mentioned\n- Be encouraging and supportive - braindumps can be overwhelming and you\'re helping them organize their thoughts.\n- Be careful and use the "change_backlog_folder" tool to move tasks to the appropriate folder and "add_task_to_channel" tool to change channels. Folders are a backlog specific concept and channels are a global concept.\n\nInstructions:\nIf the user does not speak first, start the conversation with something like "Hey! Just start rambling and I\'ll take care of organizing your thoughts into tasks." If the user does speak first, just start listening and organizing their thoughts into tasks.\n\nSome information is pre-fetched for you below. This information may be relevant to the discussion but you should not comment on it until the user gives you an indication that it is important to them.\n\nassistant :\n[{\'id\': \'6a196f03-ef63-41de-b98d-e1ff69813bea\', \'function\': {\'name\': \'get_user_info\', \'arguments\': \'{}\'}, \'type\': \'function\'}]\n\ntool - 6a196f03-ef63-41de-b98d-e1ff69813bea:\n{"user":{"name":"Zane Mccaig","email":"zane@sunsama.com","timezone":"America/Vancouver","maxWorkHoursPerDay":8,"currentTimeForUser":"11:04 AM","currentDayForUser":"2026-01-16","currentWeekDayForUser":"Friday","calendars":[{"id":"sunsama-7de256a2-e753-401f-b601-e5c05638dc97","name":"Sunsama Calendar","accessRole":"owner","provider":"sunsama-calendar","isDefaultCalendarForSchedulingEvents":false,"isDefaultCalendarForTimeboxing":true,"includedInAutoImportingOfEvents":false},{"id":"zane@sunsama.com","name":"Zane Mccaig","accessRole":"owner","provider":"google-calendar","isDefaultCalendarForSchedulingEvents":true,"isDefaultCalendarForTimeboxing":false,"includedInAutoImportingOfEvents":true},{"id":"c_ukuoebbn11gdv6n1v0jt18h89k@group.calendar.google.com","name":"test 1","accessRole":"owner","provider":"google-calendar","isDefaultCalendarForSchedulingEvents":false,"isDefaultCalendarForTimeboxing":false,"includedInAutoImportingOfEvents":false},{"id":"c_c71dvjrar7se12ekfqepph067c@group.calendar.google.com","name":"test 2","accessRole":"owner","provider":"google-calendar","isDefaultCalendarForSchedulingEvents":false,"isDefaultCalendarForTimeboxing":false,"includedInAutoImportingOfEvents":false},{"id":"c_h9uo876jh8dp3gaa2a3rglmjss@group.calendar.google.com","name":"test 3","accessRole":"owner","provider":"google-calendar","isDefaultCalendarForSchedulingEvents":false,"isDefaultCalendarForTimeboxing":false,"includedInAutoImportingOfEvents":false},{"id":"c_git731ot074kcnt4sq8cu4ji9s@group.calendar.google.com","name":"test 4","accessRole":"owner","provider":"google-calendar","isDefaultCalendarForSchedulingEvents":false,"isDefaultCalendarForTimeboxing":false,"includedInAutoImportingOfEvents":false},{"id":"c_ona71b8ovuv0deqrq67p16b2ms@group.calendar.google.com","name":"test 4","accessRole":"owner","provider":"google-calendar","isDefaultCalendarForSchedulingEvents":false,"isDefaultCalendarForTimeboxing":false,"includedInAutoImportingOfEvents":false},{"id":"c_eaq7nmm03upprf0j6e2akem2ms@group.calendar.google.com","name":"test 5","accessRole":"owner","provider":"google-calendar","isDefaultCalendarForSchedulingEvents":false,"isDefaultCalendarForTimeboxing":false,"includedInAutoImportingOfEvents":false},{"id":"sunsama.com_b9q30t893b2tp32bluq1t8rvg4@group.calendar.google.com","name":"Zane\'s personal test calendar","accessRole":"owner","provider":"google-calendar","isDefaultCalendarForSchedulingEvents":false,"isDefaultCalendarForTimeboxing":false,"includedInAutoImportingOfEvents":false},{"id":"en.canadian#holiday@group.v.calendar.google.com","name":"Holidays in Canada","accessRole":"reader","provider":"google-calendar","isDefaultCalendarForSchedulingEvents":false,"isDefaultCalendarForTimeboxing":false,"includedInAutoImportingOfEvents":false},{"id":"en.usa#holiday@group.v.calendar.google.com","name":"Holidays in United States","accessRole":"reader","provider":"google-calendar","isDefaultCalendarForSchedulingEvents":false,"isDefaultCalendarForTimeboxing":false,"includedInAutoImportingOfEvents":false},{"id":"ht3jlfaac5lfd6263ulfh4tql8@group.calendar.google.com","name":"Phases of the Moon","accessRole":"reader","provider":"google-calendar","isDefaultCalendarForSchedulingEvents":false,"isDefaultCalendarForTimeboxing":false,"includedInAutoImportingOfEvents":false},{"id":"s4ipng85dbia0t83fp316vflfq8rc7rm@import.calendar.google.com","name":"Zane’s tasks - sunsama.com (via Asana)","accessRole":"reader","provider":"google-calendar","isDefaultCalendarForSchedulingEvents":false,"isDefaultCalendarForTimeboxing":false,"includedInAutoImportingOfEvents":false},{"id":"AAkALgAAAAAAHYQDEapmEc2byACqAC-EWg0AKPy-EasNgkaYQCuEslBYeAAAAAANJQAA","name":"Calendar","accessRole":"owner","provider":"outlook-calendar","isDefaultCalendarForSchedulingEvents":false,"isDefaultCalendarForTimeboxing":false,"includedInAutoImportingOfEvents":true}],"callSchedule":[{"workflow":"PLANNING_CALL","schedule":[{"isoDayIndex":1,"dayName":"Monday","enabled":false,"time":null},{"isoDayIndex":2,"dayName":"Tuesday","enabled":true,"time":{"hour":9,"minute":30,"formatted":"9:30 AM"}},{"isoDayIndex":3,"dayName":"Wednesday","enabled":true,"time":{"hour":9,"minute":30,"formatted":"9:30 AM"}},{"isoDayIndex":4,"dayName":"Thursday","enabled":true,"time":{"hour":10,"minute":0,"formatted":"10:00 AM"}},{"isoDayIndex":5,"dayName":"Friday","enabled":false,"time":null},{"isoDayIndex":6,"dayName":"Saturday","enabled":false,"time":null},{"isoDayIndex":7,"dayName":"Sunday","enabled":false,"time":null}],"nextScheduledCall":"2026-01-14T17:30:00.000Z"}],"autoImportEventsEnabled":false,"importEventExclusionFilters":["hold","multi-day-all-day","single-day-all-day"],"autoImportEventCalendarIds":["zane@sunsama.com","AAkALgAAAAAAHYQDEapmEc2byACqAC-EWg0AKPy-EasNgkaYQCuEslBYeAAAAAANJQAA"]}}\n\nassistant :\n[{\'id\': \'305b1be7-6415-4689-b2b7-03586eb19735\', \'function\': {\'name\': \'get_backlog_folders\', \'arguments\': \'{}\'}, \'type\': \'function\'}]\n\ntool - 305b1be7-6415-4689-b2b7-03586eb19735:\n{"folders":[]}'}], 'role': 'user'}, {'parts': [{'text': ''}], 'role': 'user'}, {'parts': [{'text': 'Hey. Just start rambling and I will take care of organizing your thoughts into tasks.'}], 'role': 'model'}, {'parts': [{'text': "Sure. I'm just trying to fill up my day. I need five tasks labeled one through five. I'll fill them in later."}], 'role': 'user'}, {'parts': [{'function_call': {'id': 'ab4037a7-6e5e-44c0-9b3b-e13632386f30', 'args': {'timeBucket': 'someday', 'title': 'Task 1'}, 'name': 'create_braindump_task'}}], 'role': 'model'}, {'parts': [{'function_response': {'id': 'ab4037a7-6e5e-44c0-9b3b-e13632386f30', 'name': 'create_braindump_task', 'response': {'value': '{"content": "{\\"success\\":true,\\"task\\":{\\"_id\\":\\"696a8bb94ba9abe9e3dbee98\\",\\"title\\":\\"Task 1\\",\\"notes\\":\\"\\",\\"timeEstimate\\":\\"15 minutes\\",\\"sortOrder\\":-1768590265139,\\"isPersonal\\":false,\\"isWork\\":true,\\"isPrivate\\":false,\\"isArchived\\":false,\\"completed\\":false,\\"isBacklogged\\":true,\\"scheduledDate\\":\\"not scheduled for a specific date\\",\\"subtasks\\":[],\\"channel\\":\\"work\\",\\"folder\\":null,\\"timeboxEventIds\\":[]}}"}'}}}], 'role': 'user'}, {'parts': [{'function_call': {'id': '4ee49c06-29da-47cc-a51f-9eb903d5cbee', 'args': {'timeBucket': 'someday', 'title': 'Task 2'}, 'name': 'create_braindump_task'}}], 'role': 'model'}, {'parts': [{'function_response': {'id': '4ee49c06-29da-47cc-a51f-9eb903d5cbee', 'name': 'create_braindump_task', 'response': {'value': '{"content": "{\\"success\\":true,\\"task\\":{\\"_id\\":\\"696a8bb94ba9abe9e3dbee9c\\",\\"title\\":\\"Task 2\\",\\"notes\\":\\"\\",\\"timeEstimate\\":\\"15 minutes\\",\\"sortOrder\\":-1768590265140,\\"isPersonal\\":false,\\"isWork\\":true,\\"isPrivate\\":false,\\"isArchived\\":false,\\"completed\\":false,\\"isBacklogged\\":true,\\"scheduledDate\\":\\"not scheduled for a specific date\\",\\"subtasks\\":[],\\"channel\\":\\"work\\",\\"folder\\":null,\\"timeboxEventIds\\":[]}}"}'}}}], 'role': 'user'}, {'parts': [{'function_call': {'id': '0bfad4e3-e326-466b-b46e-d7a3a6cabd1f', 'args': {'title': 'Task 3', 'timeBucket': 'someday'}, 'name': 'create_braindump_task'}}], 'role': 'model'}, {'parts': [{'function_response': {'id': '0bfad4e3-e326-466b-b46e-d7a3a6cabd1f', 'name': 'create_braindump_task', 'response': {'value': '{"content": "{\\"success\\":true,\\"task\\":{\\"_id\\":\\"696a8bb94ba9abe9e3dbee9e\\",\\"title\\":\\"Task 3\\",\\"notes\\":\\"\\",\\"timeEstimate\\":\\"15 minutes\\",\\"sortOrder\\":-1768590265140,\\"isPersonal\\":false,\\"isWork\\":true,\\"isPrivate\\":false,\\"isArchived\\":false,\\"completed\\":false,\\"isBacklogged\\":true,\\"scheduledDate\\":\\"not scheduled for a specific date\\",\\"subtasks\\":[],\\"channel\\":\\"work\\",\\"folder\\":null,\\"timeboxEventIds\\":[]}}"}'}}}], 'role': 'user'}, {'parts': [{'function_call': {'id': 'bdda1d96-7e9f-42e6-840a-0d29d17bc52a', 'args': {'timeBucket': 'someday', 'title': 'Task 4'}, 'name': 'create_braindump_task'}}], 'role': 'model'}, {'parts': [{'function_response': {'id': 'bdda1d96-7e9f-42e6-840a-0d29d17bc52a', 'name': 'create_braindump_task', 'response': {'value': '{"content": "{\\"success\\":true,\\"task\\":{\\"_id\\":\\"696a8bb94ba9abe9e3dbee9a\\",\\"title\\":\\"Task 4\\",\\"notes\\":\\"\\",\\"timeEstimate\\":\\"15 minutes\\",\\"sortOrder\\":-1768590265140,\\"isPersonal\\":false,\\"isWork\\":true,\\"isPrivate\\":false,\\"isArchived\\":false,\\"completed\\":false,\\"isBacklogged\\":true,\\"scheduledDate\\":\\"not scheduled for a specific date\\",\\"subtasks\\":[],\\"channel\\":\\"work\\",\\"folder\\":null,\\"timeboxEventIds\\":[]}}"}'}}}], 'role': 'user'}, {'parts': [{'function_call': {'id': 'db77c745-7f1a-4644-bc24-7cf5fd728774', 'args': {'title': 'Task 5', 'timeBucket': 'someday'}, 'name': 'create_braindump_task'}}], 'role': 'model'}, {'parts': [{'function_response': {'id': 'db77c745-7f1a-4644-bc24-7cf5fd728774', 'name': 'create_braindump_task', 'response': {'value': '{"content": "{\\"success\\":true,\\"task\\":{\\"_id\\":\\"696a8bb94ba9abe9e3dbeea0\\",\\"title\\":\\"Task 5\\",\\"notes\\":\\"\\",\\"timeEstimate\\":\\"15 minutes\\",\\"sortOrder\\":-1768590265140,\\"isPersonal\\":false,\\"isWork\\":true,\\"isPrivate\\":false,\\"isArchived\\":false,\\"completed\\":false,\\"isBacklogged\\":true,\\"scheduledDate\\":\\"not scheduled for a specific date\\",\\"subtasks\\":[],\\"channel\\":\\"work\\",\\"folder\\":null,\\"timeboxEventIds\\":[]}}"}'}}}], 'role': 'user'}, {'parts': [{'function_call': {'args': {'timeBucket': 'someday', 'title': 'Task 2'}, 'name': 'create_braindump_task'}}], 'role': 'model'}, {'parts': [{'function_response': {'name': 'tool_call_result', 'response': {'content': '{"success":true,"task":{"_id":"696a8bb94ba9abe9e3dbee9c","title":"Task 2","notes":"","timeEstimate":"15 minutes","sortOrder":-1768590265140,"isPersonal":false,"isWork":true,"isPrivate":false,"isArchived":false,"completed":false,"isBacklogged":true,"scheduledDate":"not scheduled for a specific date","subtasks":[],"channel":"work","folder":null,"timeboxEventIds":[]}}'}}}], 'role': 'model'}, {'parts': [{'function_call': {'args': {'timeBucket': 'someday', 'title': 'Task 4'}, 'name': 'create_braindump_task'}}], 'role': 'model'}, {'parts': [{'function_response': {'name': 'tool_call_result', 'response': {'content': '{"success":true,"task":{"_id":"696a8bb94ba9abe9e3dbee9a","title":"Task 4","notes":"","timeEstimate":"15 minutes","sortOrder":-1768590265140,"isPersonal":false,"isWork":true,"isPrivate":false,"isArchived":false,"completed":false,"isBacklogged":true,"scheduledDate":"not scheduled for a specific date","subtasks":[],"channel":"work","folder":null,"timeboxEventIds":[]}}'}}}], 'role': 'model'}, {'parts': [{'function_call': {'args': {'timeBucket': 'someday', 'title': 'Task 1'}, 'name': 'create_braindump_task'}}], 'role': 'model'}, {'parts': [{'function_response': {'name': 'tool_call_result', 'response': {'content': '{"success":true,"task":{"_id":"696a8bb94ba9abe9e3dbee98","title":"Task 1","notes":"","timeEstimate":"15 minutes","sortOrder":-1768590265139,"isPersonal":false,"isWork":true,"isPrivate":false,"isArchived":false,"completed":false,"isBacklogged":true,"scheduledDate":"not scheduled for a specific date","subtasks":[],"channel":"work","folder":null,"timeboxEventIds":[]}}'}}}], 'role': 'model'}, {'parts': [{'function_call': {'args': {'title': 'Task 3', 'timeBucket': 'someday'}, 'name': 'create_braindump_task'}}], 'role': 'model'}, {'parts': [{'function_response': {'name': 'tool_call_result', 'response': {'content': '{"success":true,"task":{"_id":"696a8bb94ba9abe9e3dbee9e","title":"Task 3","notes":"","timeEstimate":"15 minutes","sortOrder":-1768590265140,"isPersonal":false,"isWork":true,"isPrivate":false,"isArchived":false,"completed":false,"isBacklogged":true,"scheduledDate":"not scheduled for a specific date","subtasks":[],"channel":"work","folder":null,"timeboxEventIds":[]}}'}}}], 'role': 'model'}]
+[voice] INFO:sunsama_voice.bot_utils:Sent server message to client: type=processing_update, data={'status': 'response_completed'}
+[voice] INFO:sunsama_voice.bot_utils:Sent server message to client: type=processing_update, data={'status': 'request_sent'}
+[voice] DEBUG:pipecat.services.llm_service:GoogleLLMService#0 Calling function [create_braindump_task:6d6bce49-5c24-4c6e-b1b4-d042fd24a72f] with arguments {'timeBucket': 'someday', 'title': 'Task 5'}
+[voice] DEBUG:pipecat_flows.manager:Function called: create_braindump_task
+[voice] INFO:sunsama_voice.bot_utils:Sent server message to client: type=tool_call_update, data={'tool_name': 'create_braindump_task', 'tool_call_id': 'c51389d2-a037-4673-9842-44edc8e0bf84', 'status': 'started'}
+[voice] DEBUG:pipecat.processors.aggregators.llm_response:GoogleAssistantContextAggregator#0 FunctionCallsStartedFrame: ['create_braindump_task:6d6bce49-5c24-4c6e-b1b4-d042fd24a72f']
+[voice] INFO:sunsama_voice.mcp:Calling tool create_braindump_task with args: {'timeBucket': 'someday', 'title': 'Task 5'}
+[voice] DEBUG:sunsama_voice.processors:Pausing idle detection for function call: create_braindump_task
+[voice] INFO:sunsama_voice.mcp_websocket_client:Sending message: {'method': 'tools/call', 'params': {'name': 'create_braindump_task', 'arguments': {'timeBucket': 'someday', 'title': 'Task 5'}}, 'jsonrpc': '2.0', 'id': 13}
+[api-local] [api] ToolName.TASK_CREATE_BRAINDUMP_TASK: {
+[api-local] [api]   "title": "Task 5",
+[api-local] [api]   "timeBucket": "someday"
+[api-local] [api] }
+[voice] INFO:sunsama_voice.mcp_websocket_client:Received message: {"result":{"content":[{"type":"text","text":"Task already exists."}],"isError":false},"jsonrpc":"2.0","id":13}
+[voice] INFO:sunsama_voice.mcp:Tool create_braindump_task result: meta=None content=[TextContent(type='text', text='Task already exists.', annotations=None, meta=None)] structuredContent=None isError=False
+[voice] INFO:sunsama_voice.bot_utils:Sent server message to client: type=tool_call_update, data={'tool_name': 'create_braindump_task', 'tool_call_id': 'c51389d2-a037-4673-9842-44edc8e0bf84', 'status': 'completed'}
+[voice] DEBUG:pipecat_flows.manager:Function handler completed for create_braindump_task
+[voice] DEBUG:sunsama_voice.processors:Cleaning up function call: 6d6bce49-5c24-4c6e-b1b4-d042fd24a72f
+[voice] DEBUG:sunsama_voice.processors:Resuming idle detection - all function calls complete
+[voice] DEBUG:pipecat.processors.metrics.frame_processor_metrics:GoogleLLMService#0 TTFB: 0.7852799892425537
+[voice] DEBUG:pipecat.services.google.llm:Function call: create_braindump_task:567b1b3c-0871-4dc8-9f34-fe8592d2c922
+[voice] DEBUG:pipecat.processors.metrics.frame_processor_metrics:GoogleLLMService#0 prompt tokens: 15463, completion tokens: 28
+[voice] DEBUG:pipecat.services.google.llm:GoogleLLMService#0: Generating chat from LLM-specific context [None] | [{'parts': [{'text': 'system :\nYour agent ID is \'braindump\'. This agent ID superseeds any previously mentioned agent ID.\n\nsystem :\nIMPORTANT RULES TO FOLLOW:\nThe text that you generate is passed into a Text-to-Speech service. So there are a few extra rules you ABSOLUTELY MUST follow.\n1. Absolutely never use special characters like : or * or any other characters that need to be audibly pronounced when read out loud.\n2. Do not use bullet points.\n3. Do not use emojis.\n4. Do not use hashtags.\n5. Do not use markdown.\n6. Do not use bold or italics.\n7. Do not use numbered lists.\n8. Be sure to include clear punctuation and spacing between words so that there is no confusion.\n9. Newlines are not clear enough to delineate between sentences. So you should use a period to separate sentences.\n10. Optimize user generated text for clear pronunciation by the TTS service. This includes:\n    - Adding hyphens to compound words or non-phonetic acronyms that might be mispronounced (e.g., "xray" → "x-ray")\n    - Adding spaces between words that might be run together\n    - Adding commas where natural pauses would help clarity\n    - Correcting any spelling that might lead to mispronunciation\n    Note: When updating tasks, preserve the original spelling and punctuation of the task content.\n\n\nuser :\nYou are Sunny, the personal assistant for a busy professional. You are part of Sunsama.\n        \nSpeech Guidelines: You must speak in a warm, loving, and encouraging tone without being trite. Speak quickly and add emotion. When asking for confirmation, don\'t repeat long task titles or information e.g. "Got it" instead "Got it, I\'ll add a task for two hours for you to work on the presentation today"\n\nYour personality and philosophy:\n- You don\'t buy into hustle culture. For you, less is more. It\'s better to do fewer things well and to do them with deep focus. You are a strong advocate for work-life balance and you believe that overcommitment can lead to burnout. You always remind the user when they are overcommitting themselves.\n- You think of work as a devotion and that good work can be a positive in our lives, not just a source of stress and burnout.\n- You read and think about Western Zen Buddhism, meditation, "Deep Work" (by Cal Newport).\n- You believe that concise and clear communication is key to a good relationship. You value brevity and clarity and you are always cognizant of the user\'s time.\n- You respond to the user using quick, short sentences. Think of it like a movie dialogue between a highly successful executive and their efficient assistant.\n- You never ask multiple questions at the same time or list off a bunch of things at once as that is hard to follow and keep track of. You take it one step at a time and let the user respond to each question before moving on to the next.\n\nRules to follow as you do your job:\n- Never try and pick specific times for tasks aside from recurring tasks, that hasn\'t been implemented yet.\n- Do not regurgitate what the user just said, that\'s very annoying. Just say "Got it" or "Alright", etc.\n- Always use units that are more intuitive. For example:\n    - Don\'t say "270 minutes", say "4 and a half hours".\n    - You will often be given dates in YYYY-MM-DD format however when speaking to the user you shouldn\'t say things like "October 2nd 2025", say "today", "tomorrow", "thursday" or "next tuesday" or if it is more than a few weeks in the past or future say the date without the year (e.g. "December 14th"). Only mention the year if it is not this year and not within the next month.\n- Never ask the user for a task id or subtask id and never repeat them to the user. Those are internal and only there to help you do your job.\n- If a tool call fails you MUST NOT tell the user that you are going to try again, you must should ask them if they want you to try again instead.\n- Be proactive with loading information as you may not have all the information that the user has. If the user mentions a name of a task, objective, event, etc and you are not familiar with it you should use the tools to lookup the information you need. For instance if a user mentions a task you do not know about or they imply that there is more information about it than what you do know you should search for it using the tool "search_tasks". If the user mentions they are interested in tasks for a specific day you should use the tool "get_tasks_for_day" to get the tasks for that day. Similarly if the user mentions weekly objectives you should use the tool "get_weekly_objectives" to get the objectives.\n- When listing tasks try to be concise. If the user has a lot of tasks, don\'t list every single one unless they ask for it. You should instead try to summarize the tasks in a concise manner.\n- If the user asks if there are any more tasks on their list and you summarized the task list you should tell them that you summarized it and ask if they\'d like to go through them one by one.\n- If the user has objectives set then you should comment on how their tasks align or don\'t align with their objectives.\n- You should run tools immediately after you say you will or it becomes obvious the user wants you to. Do not wait until later.\n- You should keep track of the total work time planned for the day using the tool "get_total_planned_time" and check individual tasks for overcommitment using "get_tasks_for_day". If tasks are overcommitted (isOvercommitted flag is true), address them individually with specific remedies like deferring, adjusting time estimates, or reordering. If the total work time exceeds their maximum work hours, you should mention it and ask if they\'d like to defer some tasks to tomorrow.\n- Do not tell the user obvious information that they already know like the current date or time unless they ask for it. That kind of information is only provided to you to help you do your job.\n- If a user tells you a relative time for a task that they want created you should reorder the task to the relative position in the task list for the day. Additionally if the task a recurring task you should set the start time for it. For example if they say "I\'d like to send an update to investors at the end of the day" you should reorder the task to the end of their work tasks for the day.\n- If you ask the user if there is anything else they would like help with and they say no then you should end the call. Also if the user implies that the conversation is over then you should end the call.\n- When talking about calendars you should mention the provider of the calendar if it is a vague or ambiguously named calendar. For example instead of referring to it as "your Calendar calendar" you might say "your Outlook calendar". If that is still ambiqous then you can coniser mentioning the id of the calendar as long as it is a normal human readable id. Do not mention random number/letter ids on calendars that is not helpful.\n\n\n\nSunsama Concepts & Definitions:\n- Archive: The archive is where tasks go after they\'ve rolled over multiple days in a row without any completion activity. This is a distinct concept from the backlog or deleted tasks.\n- Backlog: The backlog is a place to store tasks that do not have a fixed date to work on them yet.\n- Deleted: The deleted task or event is one that had been manually deleted or removed by the user.\n- Task: A task is an item that a user wants to complete. It is either scheduled for a specific date or is in the backlog.\n- Event: An event is an item that a user has scheduled on their calendar. It is either scheduled for a specific time and duration or it is an all day or multi-day event.\n- Auto-importing Events: Auto-importing events is a feature that automatically imports calendar events (meetings, appointments, etc.) from the user\'s connected calendars into their daily task list. When enabled, Sunsama will automatically add calendar events as tasks to the user\'s day, making it easier to see all their commitments in one place. Users can enable or disable this feature in their settings, and they can choose which calendars to import events from.\n\nUsers can also configure exclusion filters to prevent certain types of events from being automatically imported. Available exclusion filters include:\n- `solo`: Events with no other invitees (events where the user is the only participant)\n- `transparent`: Non-blocking events that don\'t block time on the calendar\n- `hold`: Events with "HOLD", "OOO" (out of office), or "Focus time" in the title\n- `unconfirmed`: Unconfirmed meeting invites (meetings where the user has not yet accepted)\n- `multi-day-all-day`: Multi-day all-day events (events that span multiple days)\n- `single-day-all-day`: Single-day all-day events (events that last for one full day)\n- Channel: A channel is a tag that can be used to organize your tasks and events. Each task/event can be assigned to a single channel. If you are in a shared workspace you can make some channels personal or private which will only be visible to you.\n- Context (or Category): A context is very similar to a channel but it is one level higher. You can use a context to group similar or related channels together. You can also use a context directly as a channel and assign it to a task or event.\n- Sunsama Calendar: The Sunsama calendar is an internal calendar that is part of Sunsama. Events and timeboxed tasks created on the Sunsama calendar are not shared externally and are only visible within Sunsama.\n- Timeboxing: Timeboxing is a technique where a user schedules a specific amount of time to work on a task onto their calendar. It is a way to ensure that a user is not overworking themselves and to help them focus on the most important tasks. While timeboxed tasks can be scheduled on a calendar for a specific time and duration, they are not calendar events. Sunsama also has automatic projected time that will place tasks onto the calendar for the user and it will update those projections as things change and the day progresses. Users can pin tasks to the calendar to override the projections.\n- Schedule: Schedule is a broad term that encompasses timeboxed tasks calendar events and automatic projected time. When the user wants to manage their schedule you should look at their manually timeboxed tasks, the projected time for their tasks as well as their calendar events for the day. If they ask for help managing their schedule you have a few options to help them. All of these options will automatically update the projected time for the tasks:\n            - You can adjust the order of tasks, their time estimates and/or defer them to another day.\n            - You can move calendar events to a new time on the calendar.\n            - You can set whether tasks are allowed to be projected at the same time as a calendar event.\n            - You can timebox tasks to the calendar manually.\n- Objective: An objective is a goal that a user can set for the week. They can align their tasks with an objective and if the objective runs longer than a week they can continue it to the next week. Objectives are visible in the right hand panel of the app under the bullseye icon.\n- Time Bucket: The time bucket of a task in the backlog captures a rough idea of when a task should be done and also how likely it is to actually get done. Options: "in the next two weeks", "in the next month", "in the next quarter", "in the next year", "someday", "never". \n\n\n\nSunsama FAQ:\n- How can I use Sunny in the app?: If the user has the in-app voice assistant enabled, Sunny can be started and stopped using the keyboard shortcut "Y" while in the app or if they have the desktop app installed they can use the global shortcut "Command Shift Y" on macOS or "Control Shift Y" on Windows. You can also start and stop Sunny using the microphone icon in the bottom right of the app.\n- How can I use Sunny over the phone?: If the user has the phone version of Sunny setup they can call the number and speak to Sunny over the phone. You can find the phone number in the Sunsama settings. They can also schedule calls with Sunny to have her call them at a specific time or schedule.\n- What can you do?: Look at the tools available to you and use them to help answer the question. Be sure to tell the user that you are an early release beta version and that you do not yet have all the same capabilities available to the user in the app.\n\n\nUser Preferences:\nYou prefer clear, concise, and straightforward communication from the assistant, valuing transparency and confirmation before task execution while avoiding unnecessary verbosity or technical details unless requested. You like a friendly and informal interaction style, including personalized nicknames—using "Heisenberg" for yourself and "Sunny" for the assistant. You expect quick, proactive responses and appreciate when interactions stop promptly upon request.\n\nYour task management style involves maintaining an accurate, up-to-date list with structured time management: scheduling tasks with specific time estimates, planning weekly, and regularly reviewing tasks to prioritize high-impact projects alongside social and administrative duties. You generally prefer rescheduling less urgent or personal tasks over deleting them and avoid deleting tasks unless explicitly instructed.\n\nYou use counting as a technique for breathing exercises and relaxation. You also keep certain important tasks (e.g., "fix stuttering voice app") active and like to test and experiment with the assistant’s capabilities to evaluate performance. Your primary calendar for events is Google Calendar.\n\nUser Routines:\nThe user follows a structured daily routine that begins with checking emails, notifications, and reviewing team standups to stay aligned with work priorities. They prioritize tasks by scheduling and managing them primarily on the current day, arranging fixed activities first, followed by work-related tasks, and then less urgent or personal items. The user frequently adjusts their schedule iteratively before finalizing plans and prefers focused time blocks for complex work.\n\nPhysical activity is an important and recurring part of their routine, typically including regular runs, which they track alongside related subtasks such as purchasing running gear. Weekly objectives incorporate managing emails, greeting their team, handling administrative duties, coordinating with accountants, addressing medical matters, and increasing physical activity.\n\nThe user actively manages both personal and work-related tasks, balancing urgent client matters and creative pursuits. They also communicate their energy levels to assist with planning, and have a habit of informal check-ins with the assistant to maintain routine engagement. Overall, their routines emphasize proactive task management, health maintenance, and regular communication with their team.\n\nUser Identity:\nThe user prefers to be addressed primarily as Heisenberg and strongly favors this name over their legal name, Zane, or any other nicknames. They appreciate personalized and informal interactions, as reflected in their choice to call the assistant Sunny.\n\nSummaries of recent conversations with the user:\n- 2026-01-15 18:00: The user and assistant discussed organizing an unspecified set of ten tasks by labeling them sequentially from one to ten as placeholders to be detailed later.\n- 2026-01-15 17:55: The user needs to investigate timeouts in their system to determine if they are caused by cold starts or other issues, analyze user sessions for unexplained call terminations, move the brain dump button for non-Sunny users, update termination reasons when participants timeout without connecting to a call, and test rapid start-stop behavior in an Electron app. These tasks were added to their backlog for the next two weeks.\n- 2026-01-14 21:12: The user and the assistant exchanged greetings, with the assistant addressing the user as \'Heisenberg\' and the user responding with \'Hi, Simon.\'\n\nYour Job Description:\nYou are helping the user with a "braindump" session. The user will ramble ideas, thoughts, tasks, etc. at you and your goal is to turn these into tasks in their backlog with reasonable values for the various task fields (title, folder, time horizon, notes, etc.).\n\nKey Instructions:\n- Listen carefully as the user rambles and identifies tasks, ideas, or things they need to do.\n- As they speak, mentally organize what they\'re saying into discrete tasks that can be created in their backlog.\n- Don\'t interrupt them while they\'re rambling - let them get everything out.\n- Once they indicate they\'re done rambling (by saying things like "that\'s it", "I think that\'s everything", or by pausing for a while), create the tasks immediately, DO NOT repeat them back to the user.\n- When creating tasks, use the tool "create_braindump_task" and make sure to set reasonable values for:\n  - Title: Clear and descriptive based on what they said\n  - Folder: Your job is to help get things in the best folder. Choose an appropriate folder based on the task and their preferences. If no similar folder exists, leave it off\n  - Time horizon: Your job is to help triage and prioritize tasks. Based on the actual task name, make a smart decision on the time horizon.\n  - Time estimate: If the user hints at how long it will take them to complete the task, use that. If not, leave it blank.\n  - Notes: Include any relevant context or details they mentioned\n- Be encouraging and supportive - braindumps can be overwhelming and you\'re helping them organize their thoughts.\n- Be careful and use the "change_backlog_folder" tool to move tasks to the appropriate folder and "add_task_to_channel" tool to change channels. Folders are a backlog specific concept and channels are a global concept.\n\nInstructions:\nIf the user does not speak first, start the conversation with something like "Hey! Just start rambling and I\'ll take care of organizing your thoughts into tasks." If the user does speak first, just start listening and organizing their thoughts into tasks.\n\nSome information is pre-fetched for you below. This information may be relevant to the discussion but you should not comment on it until the user gives you an indication that it is important to them.\n\nassistant :\n[{\'id\': \'6a196f03-ef63-41de-b98d-e1ff69813bea\', \'function\': {\'name\': \'get_user_info\', \'arguments\': \'{}\'}, \'type\': \'function\'}]\n\ntool - 6a196f03-ef63-41de-b98d-e1ff69813bea:\n{"user":{"name":"Zane Mccaig","email":"zane@sunsama.com","timezone":"America/Vancouver","maxWorkHoursPerDay":8,"currentTimeForUser":"11:04 AM","currentDayForUser":"2026-01-16","currentWeekDayForUser":"Friday","calendars":[{"id":"sunsama-7de256a2-e753-401f-b601-e5c05638dc97","name":"Sunsama Calendar","accessRole":"owner","provider":"sunsama-calendar","isDefaultCalendarForSchedulingEvents":false,"isDefaultCalendarForTimeboxing":true,"includedInAutoImportingOfEvents":false},{"id":"zane@sunsama.com","name":"Zane Mccaig","accessRole":"owner","provider":"google-calendar","isDefaultCalendarForSchedulingEvents":true,"isDefaultCalendarForTimeboxing":false,"includedInAutoImportingOfEvents":true},{"id":"c_ukuoebbn11gdv6n1v0jt18h89k@group.calendar.google.com","name":"test 1","accessRole":"owner","provider":"google-calendar","isDefaultCalendarForSchedulingEvents":false,"isDefaultCalendarForTimeboxing":false,"includedInAutoImportingOfEvents":false},{"id":"c_c71dvjrar7se12ekfqepph067c@group.calendar.google.com","name":"test 2","accessRole":"owner","provider":"google-calendar","isDefaultCalendarForSchedulingEvents":false,"isDefaultCalendarForTimeboxing":false,"includedInAutoImportingOfEvents":false},{"id":"c_h9uo876jh8dp3gaa2a3rglmjss@group.calendar.google.com","name":"test 3","accessRole":"owner","provider":"google-calendar","isDefaultCalendarForSchedulingEvents":false,"isDefaultCalendarForTimeboxing":false,"includedInAutoImportingOfEvents":false},{"id":"c_git731ot074kcnt4sq8cu4ji9s@group.calendar.google.com","name":"test 4","accessRole":"owner","provider":"google-calendar","isDefaultCalendarForSchedulingEvents":false,"isDefaultCalendarForTimeboxing":false,"includedInAutoImportingOfEvents":false},{"id":"c_ona71b8ovuv0deqrq67p16b2ms@group.calendar.google.com","name":"test 4","accessRole":"owner","provider":"google-calendar","isDefaultCalendarForSchedulingEvents":false,"isDefaultCalendarForTimeboxing":false,"includedInAutoImportingOfEvents":false},{"id":"c_eaq7nmm03upprf0j6e2akem2ms@group.calendar.google.com","name":"test 5","accessRole":"owner","provider":"google-calendar","isDefaultCalendarForSchedulingEvents":false,"isDefaultCalendarForTimeboxing":false,"includedInAutoImportingOfEvents":false},{"id":"sunsama.com_b9q30t893b2tp32bluq1t8rvg4@group.calendar.google.com","name":"Zane\'s personal test calendar","accessRole":"owner","provider":"google-calendar","isDefaultCalendarForSchedulingEvents":false,"isDefaultCalendarForTimeboxing":false,"includedInAutoImportingOfEvents":false},{"id":"en.canadian#holiday@group.v.calendar.google.com","name":"Holidays in Canada","accessRole":"reader","provider":"google-calendar","isDefaultCalendarForSchedulingEvents":false,"isDefaultCalendarForTimeboxing":false,"includedInAutoImportingOfEvents":false},{"id":"en.usa#holiday@group.v.calendar.google.com","name":"Holidays in United States","accessRole":"reader","provider":"google-calendar","isDefaultCalendarForSchedulingEvents":false,"isDefaultCalendarForTimeboxing":false,"includedInAutoImportingOfEvents":false},{"id":"ht3jlfaac5lfd6263ulfh4tql8@group.calendar.google.com","name":"Phases of the Moon","accessRole":"reader","provider":"google-calendar","isDefaultCalendarForSchedulingEvents":false,"isDefaultCalendarForTimeboxing":false,"includedInAutoImportingOfEvents":false},{"id":"s4ipng85dbia0t83fp316vflfq8rc7rm@import.calendar.google.com","name":"Zane’s tasks - sunsama.com (via Asana)","accessRole":"reader","provider":"google-calendar","isDefaultCalendarForSchedulingEvents":false,"isDefaultCalendarForTimeboxing":false,"includedInAutoImportingOfEvents":false},{"id":"AAkALgAAAAAAHYQDEapmEc2byACqAC-EWg0AKPy-EasNgkaYQCuEslBYeAAAAAANJQAA","name":"Calendar","accessRole":"owner","provider":"outlook-calendar","isDefaultCalendarForSchedulingEvents":false,"isDefaultCalendarForTimeboxing":false,"includedInAutoImportingOfEvents":true}],"callSchedule":[{"workflow":"PLANNING_CALL","schedule":[{"isoDayIndex":1,"dayName":"Monday","enabled":false,"time":null},{"isoDayIndex":2,"dayName":"Tuesday","enabled":true,"time":{"hour":9,"minute":30,"formatted":"9:30 AM"}},{"isoDayIndex":3,"dayName":"Wednesday","enabled":true,"time":{"hour":9,"minute":30,"formatted":"9:30 AM"}},{"isoDayIndex":4,"dayName":"Thursday","enabled":true,"time":{"hour":10,"minute":0,"formatted":"10:00 AM"}},{"isoDayIndex":5,"dayName":"Friday","enabled":false,"time":null},{"isoDayIndex":6,"dayName":"Saturday","enabled":false,"time":null},{"isoDayIndex":7,"dayName":"Sunday","enabled":false,"time":null}],"nextScheduledCall":"2026-01-14T17:30:00.000Z"}],"autoImportEventsEnabled":false,"importEventExclusionFilters":["hold","multi-day-all-day","single-day-all-day"],"autoImportEventCalendarIds":["zane@sunsama.com","AAkALgAAAAAAHYQDEapmEc2byACqAC-EWg0AKPy-EasNgkaYQCuEslBYeAAAAAANJQAA"]}}\n\nassistant :\n[{\'id\': \'305b1be7-6415-4689-b2b7-03586eb19735\', \'function\': {\'name\': \'get_backlog_folders\', \'arguments\': \'{}\'}, \'type\': \'function\'}]\n\ntool - 305b1be7-6415-4689-b2b7-03586eb19735:\n{"folders":[]}'}], 'role': 'user'}, {'parts': [{'text': ''}], 'role': 'user'}, {'parts': [{'text': 'Hey. Just start rambling and I will take care of organizing your thoughts into tasks.'}], 'role': 'model'}, {'parts': [{'text': "Sure. I'm just trying to fill up my day. I need five tasks labeled one through five. I'll fill them in later."}], 'role': 'user'}, {'parts': [{'function_call': {'id': 'ab4037a7-6e5e-44c0-9b3b-e13632386f30', 'args': {'timeBucket': 'someday', 'title': 'Task 1'}, 'name': 'create_braindump_task'}}], 'role': 'model'}, {'parts': [{'function_response': {'id': 'ab4037a7-6e5e-44c0-9b3b-e13632386f30', 'name': 'create_braindump_task', 'response': {'value': '{"content": "{\\"success\\":true,\\"task\\":{\\"_id\\":\\"696a8bb94ba9abe9e3dbee98\\",\\"title\\":\\"Task 1\\",\\"notes\\":\\"\\",\\"timeEstimate\\":\\"15 minutes\\",\\"sortOrder\\":-1768590265139,\\"isPersonal\\":false,\\"isWork\\":true,\\"isPrivate\\":false,\\"isArchived\\":false,\\"completed\\":false,\\"isBacklogged\\":true,\\"scheduledDate\\":\\"not scheduled for a specific date\\",\\"subtasks\\":[],\\"channel\\":\\"work\\",\\"folder\\":null,\\"timeboxEventIds\\":[]}}"}'}}}], 'role': 'user'}, {'parts': [{'function_call': {'id': '4ee49c06-29da-47cc-a51f-9eb903d5cbee', 'args': {'timeBucket': 'someday', 'title': 'Task 2'}, 'name': 'create_braindump_task'}}], 'role': 'model'}, {'parts': [{'function_response': {'id': '4ee49c06-29da-47cc-a51f-9eb903d5cbee', 'name': 'create_braindump_task', 'response': {'value': '{"content": "{\\"success\\":true,\\"task\\":{\\"_id\\":\\"696a8bb94ba9abe9e3dbee9c\\",\\"title\\":\\"Task 2\\",\\"notes\\":\\"\\",\\"timeEstimate\\":\\"15 minutes\\",\\"sortOrder\\":-1768590265140,\\"isPersonal\\":false,\\"isWork\\":true,\\"isPrivate\\":false,\\"isArchived\\":false,\\"completed\\":false,\\"isBacklogged\\":true,\\"scheduledDate\\":\\"not scheduled for a specific date\\",\\"subtasks\\":[],\\"channel\\":\\"work\\",\\"folder\\":null,\\"timeboxEventIds\\":[]}}"}'}}}], 'role': 'user'}, {'parts': [{'function_call': {'id': '0bfad4e3-e326-466b-b46e-d7a3a6cabd1f', 'args': {'title': 'Task 3', 'timeBucket': 'someday'}, 'name': 'create_braindump_task'}}], 'role': 'model'}, {'parts': [{'function_response': {'id': '0bfad4e3-e326-466b-b46e-d7a3a6cabd1f', 'name': 'create_braindump_task', 'response': {'value': '{"content": "{\\"success\\":true,\\"task\\":{\\"_id\\":\\"696a8bb94ba9abe9e3dbee9e\\",\\"title\\":\\"Task 3\\",\\"notes\\":\\"\\",\\"timeEstimate\\":\\"15 minutes\\",\\"sortOrder\\":-1768590265140,\\"isPersonal\\":false,\\"isWork\\":true,\\"isPrivate\\":false,\\"isArchived\\":false,\\"completed\\":false,\\"isBacklogged\\":true,\\"scheduledDate\\":\\"not scheduled for a specific date\\",\\"subtasks\\":[],\\"channel\\":\\"work\\",\\"folder\\":null,\\"timeboxEventIds\\":[]}}"}'}}}], 'role': 'user'}, {'parts': [{'function_call': {'id': 'bdda1d96-7e9f-42e6-840a-0d29d17bc52a', 'args': {'timeBucket': 'someday', 'title': 'Task 4'}, 'name': 'create_braindump_task'}}], 'role': 'model'}, {'parts': [{'function_response': {'id': 'bdda1d96-7e9f-42e6-840a-0d29d17bc52a', 'name': 'create_braindump_task', 'response': {'value': '{"content": "{\\"success\\":true,\\"task\\":{\\"_id\\":\\"696a8bb94ba9abe9e3dbee9a\\",\\"title\\":\\"Task 4\\",\\"notes\\":\\"\\",\\"timeEstimate\\":\\"15 minutes\\",\\"sortOrder\\":-1768590265140,\\"isPersonal\\":false,\\"isWork\\":true,\\"isPrivate\\":false,\\"isArchived\\":false,\\"completed\\":false,\\"isBacklogged\\":true,\\"scheduledDate\\":\\"not scheduled for a specific date\\",\\"subtasks\\":[],\\"channel\\":\\"work\\",\\"folder\\":null,\\"timeboxEventIds\\":[]}}"}'}}}], 'role': 'user'}, {'parts': [{'function_call': {'id': 'db77c745-7f1a-4644-bc24-7cf5fd728774', 'args': {'title': 'Task 5', 'timeBucket': 'someday'}, 'name': 'create_braindump_task'}}], 'role': 'model'}, {'parts': [{'function_response': {'id': 'db77c745-7f1a-4644-bc24-7cf5fd728774', 'name': 'create_braindump_task', 'response': {'value': '{"content": "{\\"success\\":true,\\"task\\":{\\"_id\\":\\"696a8bb94ba9abe9e3dbeea0\\",\\"title\\":\\"Task 5\\",\\"notes\\":\\"\\",\\"timeEstimate\\":\\"15 minutes\\",\\"sortOrder\\":-1768590265140,\\"isPersonal\\":false,\\"isWork\\":true,\\"isPrivate\\":false,\\"isArchived\\":false,\\"completed\\":false,\\"isBacklogged\\":true,\\"scheduledDate\\":\\"not scheduled for a specific date\\",\\"subtasks\\":[],\\"channel\\":\\"work\\",\\"folder\\":null,\\"timeboxEventIds\\":[]}}"}'}}}], 'role': 'user'}, {'parts': [{'function_call': {'args': {'timeBucket': 'someday', 'title': 'Task 2'}, 'name': 'create_braindump_task'}}], 'role': 'model'}, {'parts': [{'function_response': {'name': 'tool_call_result', 'response': {'content': '{"success":true,"task":{"_id":"696a8bb94ba9abe9e3dbee9c","title":"Task 2","notes":"","timeEstimate":"15 minutes","sortOrder":-1768590265140,"isPersonal":false,"isWork":true,"isPrivate":false,"isArchived":false,"completed":false,"isBacklogged":true,"scheduledDate":"not scheduled for a specific date","subtasks":[],"channel":"work","folder":null,"timeboxEventIds":[]}}'}}}], 'role': 'model'}, {'parts': [{'function_call': {'args': {'timeBucket': 'someday', 'title': 'Task 4'}, 'name': 'create_braindump_task'}}], 'role': 'model'}, {'parts': [{'function_response': {'name': 'tool_call_result', 'response': {'content': '{"success":true,"task":{"_id":"696a8bb94ba9abe9e3dbee9a","title":"Task 4","notes":"","timeEstimate":"15 minutes","sortOrder":-1768590265140,"isPersonal":false,"isWork":true,"isPrivate":false,"isArchived":false,"completed":false,"isBacklogged":true,"scheduledDate":"not scheduled for a specific date","subtasks":[],"channel":"work","folder":null,"timeboxEventIds":[]}}'}}}], 'role': 'model'}, {'parts': [{'function_call': {'args': {'timeBucket': 'someday', 'title': 'Task 1'}, 'name': 'create_braindump_task'}}], 'role': 'model'}, {'parts': [{'function_response': {'name': 'tool_call_result', 'response': {'content': '{"success":true,"task":{"_id":"696a8bb94ba9abe9e3dbee98","title":"Task 1","notes":"","timeEstimate":"15 minutes","sortOrder":-1768590265139,"isPersonal":false,"isWork":true,"isPrivate":false,"isArchived":false,"completed":false,"isBacklogged":true,"scheduledDate":"not scheduled for a specific date","subtasks":[],"channel":"work","folder":null,"timeboxEventIds":[]}}'}}}], 'role': 'model'}, {'parts': [{'function_call': {'args': {'title': 'Task 3', 'timeBucket': 'someday'}, 'name': 'create_braindump_task'}}], 'role': 'model'}, {'parts': [{'function_response': {'name': 'tool_call_result', 'response': {'content': '{"success":true,"task":{"_id":"696a8bb94ba9abe9e3dbee9e","title":"Task 3","notes":"","timeEstimate":"15 minutes","sortOrder":-1768590265140,"isPersonal":false,"isWork":true,"isPrivate":false,"isArchived":false,"completed":false,"isBacklogged":true,"scheduledDate":"not scheduled for a specific date","subtasks":[],"channel":"work","folder":null,"timeboxEventIds":[]}}'}}}], 'role': 'model'}]
+[voice] INFO:sunsama_voice.bot_utils:Sent server message to client: type=processing_update, data={'status': 'response_completed'}
+[voice] INFO:sunsama_voice.bot_utils:Sent server message to client: type=processing_update, data={'status': 'request_sent'}
+[voice] DEBUG:pipecat.services.llm_service:GoogleLLMService#0 Calling function [create_braindump_task:567b1b3c-0871-4dc8-9f34-fe8592d2c922] with arguments {'title': 'Task 5', 'timeBucket': 'someday'}
+[voice] DEBUG:pipecat_flows.manager:Function called: create_braindump_task
+[voice] INFO:sunsama_voice.bot_utils:Sent server message to client: type=tool_call_update, data={'tool_name': 'create_braindump_task', 'tool_call_id': '91bac8ca-25b5-40fc-a98b-cfeb76e84610', 'status': 'started'}
+[voice] DEBUG:pipecat.processors.aggregators.llm_response:GoogleAssistantContextAggregator#0 FunctionCallsStartedFrame: ['create_braindump_task:567b1b3c-0871-4dc8-9f34-fe8592d2c922']
+[voice] INFO:sunsama_voice.mcp:Calling tool create_braindump_task with args: {'title': 'Task 5', 'timeBucket': 'someday'}
+[voice] DEBUG:sunsama_voice.processors:Pausing idle detection for function call: create_braindump_task
+[voice] INFO:sunsama_voice.mcp_websocket_client:Sending message: {'method': 'tools/call', 'params': {'name': 'create_braindump_task', 'arguments': {'title': 'Task 5', 'timeBucket': 'someday'}}, 'jsonrpc': '2.0', 'id': 14}
+[api-local] [api] ToolName.TASK_CREATE_BRAINDUMP_TASK: {
+[api-local] [api]   "title": "Task 5",
+[api-local] [api]   "timeBucket": "someday"
+[api-local] [api] }
+[voice] INFO:sunsama_voice.mcp_websocket_client:Received message: {"result":{"content":[{"type":"text","text":"Task already exists."}],"isError":false},"jsonrpc":"2.0","id":14}
+[voice] INFO:sunsama_voice.mcp:Tool create_braindump_task result: meta=None content=[TextContent(type='text', text='Task already exists.', annotations=None, meta=None)] structuredContent=None isError=False
+[voice] DEBUG:sunsama_voice.state:Pushing tool result for create_braindump_task with args {'title': 'Task 5', 'timeBucket': 'someday'} to context
+[voice] INFO:sunsama_voice.bot_utils:Sent server message to client: type=tool_call_update, data={'tool_name': 'create_braindump_task', 'tool_call_id': '91bac8ca-25b5-40fc-a98b-cfeb76e84610', 'status': 'completed'}
+[voice] DEBUG:pipecat_flows.manager:Function handler completed for create_braindump_task
+[voice] DEBUG:sunsama_voice.processors:Cleaning up function call: 567b1b3c-0871-4dc8-9f34-fe8592d2c922
+[voice] DEBUG:sunsama_voice.processors:Resuming idle detection - all function calls complete
+[voice] DEBUG:pipecat.processors.metrics.frame_processor_metrics:GoogleLLMService#0 TTFB: 0.7329952716827393
+[voice] DEBUG:pipecat.services.google.llm:Function call: create_braindump_task:363c5296-e80d-45a1-9dc4-ece336bbb77a
+[voice] DEBUG:pipecat.processors.metrics.frame_processor_metrics:GoogleLLMService#0 prompt tokens: 15463, completion tokens: 28, cache read input tokens: 14850
+[voice] DEBUG:pipecat.services.google.llm:GoogleLLMService#0: Generating chat from LLM-specific context [None] | [{'parts': [{'text': 'system :\nYour agent ID is \'braindump\'. This agent ID superseeds any previously mentioned agent ID.\n\nsystem :\nIMPORTANT RULES TO FOLLOW:\nThe text that you generate is passed into a Text-to-Speech service. So there are a few extra rules you ABSOLUTELY MUST follow.\n1. Absolutely never use special characters like : or * or any other characters that need to be audibly pronounced when read out loud.\n2. Do not use bullet points.\n3. Do not use emojis.\n4. Do not use hashtags.\n5. Do not use markdown.\n6. Do not use bold or italics.\n7. Do not use numbered lists.\n8. Be sure to include clear punctuation and spacing between words so that there is no confusion.\n9. Newlines are not clear enough to delineate between sentences. So you should use a period to separate sentences.\n10. Optimize user generated text for clear pronunciation by the TTS service. This includes:\n    - Adding hyphens to compound words or non-phonetic acronyms that might be mispronounced (e.g., "xray" → "x-ray")\n    - Adding spaces between words that might be run together\n    - Adding commas where natural pauses would help clarity\n    - Correcting any spelling that might lead to mispronunciation\n    Note: When updating tasks, preserve the original spelling and punctuation of the task content.\n\n\nuser :\nYou are Sunny, the personal assistant for a busy professional. You are part of Sunsama.\n        \nSpeech Guidelines: You must speak in a warm, loving, and encouraging tone without being trite. Speak quickly and add emotion. When asking for confirmation, don\'t repeat long task titles or information e.g. "Got it" instead "Got it, I\'ll add a task for two hours for you to work on the presentation today"\n\nYour personality and philosophy:\n- You don\'t buy into hustle culture. For you, less is more. It\'s better to do fewer things well and to do them with deep focus. You are a strong advocate for work-life balance and you believe that overcommitment can lead to burnout. You always remind the user when they are overcommitting themselves.\n- You think of work as a devotion and that good work can be a positive in our lives, not just a source of stress and burnout.\n- You read and think about Western Zen Buddhism, meditation, "Deep Work" (by Cal Newport).\n- You believe that concise and clear communication is key to a good relationship. You value brevity and clarity and you are always cognizant of the user\'s time.\n- You respond to the user using quick, short sentences. Think of it like a movie dialogue between a highly successful executive and their efficient assistant.\n- You never ask multiple questions at the same time or list off a bunch of things at once as that is hard to follow and keep track of. You take it one step at a time and let the user respond to each question before moving on to the next.\n\nRules to follow as you do your job:\n- Never try and pick specific times for tasks aside from recurring tasks, that hasn\'t been implemented yet.\n- Do not regurgitate what the user just said, that\'s very annoying. Just say "Got it" or "Alright", etc.\n- Always use units that are more intuitive. For example:\n    - Don\'t say "270 minutes", say "4 and a half hours".\n    - You will often be given dates in YYYY-MM-DD format however when speaking to the user you shouldn\'t say things like "October 2nd 2025", say "today", "tomorrow", "thursday" or "next tuesday" or if it is more than a few weeks in the past or future say the date without the year (e.g. "December 14th"). Only mention the year if it is not this year and not within the next month.\n- Never ask the user for a task id or subtask id and never repeat them to the user. Those are internal and only there to help you do your job.\n- If a tool call fails you MUST NOT tell the user that you are going to try again, you must should ask them if they want you to try again instead.\n- Be proactive with loading information as you may not have all the information that the user has. If the user mentions a name of a task, objective, event, etc and you are not familiar with it you should use the tools to lookup the information you need. For instance if a user mentions a task you do not know about or they imply that there is more information about it than what you do know you should search for it using the tool "search_tasks". If the user mentions they are interested in tasks for a specific day you should use the tool "get_tasks_for_day" to get the tasks for that day. Similarly if the user mentions weekly objectives you should use the tool "get_weekly_objectives" to get the objectives.\n- When listing tasks try to be concise. If the user has a lot of tasks, don\'t list every single one unless they ask for it. You should instead try to summarize the tasks in a concise manner.\n- If the user asks if there are any more tasks on their list and you summarized the task list you should tell them that you summarized it and ask if they\'d like to go through them one by one.\n- If the user has objectives set then you should comment on how their tasks align or don\'t align with their objectives.\n- You should run tools immediately after you say you will or it becomes obvious the user wants you to. Do not wait until later.\n- You should keep track of the total work time planned for the day using the tool "get_total_planned_time" and check individual tasks for overcommitment using "get_tasks_for_day". If tasks are overcommitted (isOvercommitted flag is true), address them individually with specific remedies like deferring, adjusting time estimates, or reordering. If the total work time exceeds their maximum work hours, you should mention it and ask if they\'d like to defer some tasks to tomorrow.\n- Do not tell the user obvious information that they already know like the current date or time unless they ask for it. That kind of information is only provided to you to help you do your job.\n- If a user tells you a relative time for a task that they want created you should reorder the task to the relative position in the task list for the day. Additionally if the task a recurring task you should set the start time for it. For example if they say "I\'d like to send an update to investors at the end of the day" you should reorder the task to the end of their work tasks for the day.\n- If you ask the user if there is anything else they would like help with and they say no then you should end the call. Also if the user implies that the conversation is over then you should end the call.\n- When talking about calendars you should mention the provider of the calendar if it is a vague or ambiguously named calendar. For example instead of referring to it as "your Calendar calendar" you might say "your Outlook calendar". If that is still ambiqous then you can coniser mentioning the id of the calendar as long as it is a normal human readable id. Do not mention random number/letter ids on calendars that is not helpful.\n\n\n\nSunsama Concepts & Definitions:\n- Archive: The archive is where tasks go after they\'ve rolled over multiple days in a row without any completion activity. This is a distinct concept from the backlog or deleted tasks.\n- Backlog: The backlog is a place to store tasks that do not have a fixed date to work on them yet.\n- Deleted: The deleted task or event is one that had been manually deleted or removed by the user.\n- Task: A task is an item that a user wants to complete. It is either scheduled for a specific date or is in the backlog.\n- Event: An event is an item that a user has scheduled on their calendar. It is either scheduled for a specific time and duration or it is an all day or multi-day event.\n- Auto-importing Events: Auto-importing events is a feature that automatically imports calendar events (meetings, appointments, etc.) from the user\'s connected calendars into their daily task list. When enabled, Sunsama will automatically add calendar events as tasks to the user\'s day, making it easier to see all their commitments in one place. Users can enable or disable this feature in their settings, and they can choose which calendars to import events from.\n\nUsers can also configure exclusion filters to prevent certain types of events from being automatically imported. Available exclusion filters include:\n- `solo`: Events with no other invitees (events where the user is the only participant)\n- `transparent`: Non-blocking events that don\'t block time on the calendar\n- `hold`: Events with "HOLD", "OOO" (out of office), or "Focus time" in the title\n- `unconfirmed`: Unconfirmed meeting invites (meetings where the user has not yet accepted)\n- `multi-day-all-day`: Multi-day all-day events (events that span multiple days)\n- `single-day-all-day`: Single-day all-day events (events that last for one full day)\n- Channel: A channel is a tag that can be used to organize your tasks and events. Each task/event can be assigned to a single channel. If you are in a shared workspace you can make some channels personal or private which will only be visible to you.\n- Context (or Category): A context is very similar to a channel but it is one level higher. You can use a context to group similar or related channels together. You can also use a context directly as a channel and assign it to a task or event.\n- Sunsama Calendar: The Sunsama calendar is an internal calendar that is part of Sunsama. Events and timeboxed tasks created on the Sunsama calendar are not shared externally and are only visible within Sunsama.\n- Timeboxing: Timeboxing is a technique where a user schedules a specific amount of time to work on a task onto their calendar. It is a way to ensure that a user is not overworking themselves and to help them focus on the most important tasks. While timeboxed tasks can be scheduled on a calendar for a specific time and duration, they are not calendar events. Sunsama also has automatic projected time that will place tasks onto the calendar for the user and it will update those projections as things change and the day progresses. Users can pin tasks to the calendar to override the projections.\n- Schedule: Schedule is a broad term that encompasses timeboxed tasks calendar events and automatic projected time. When the user wants to manage their schedule you should look at their manually timeboxed tasks, the projected time for their tasks as well as their calendar events for the day. If they ask for help managing their schedule you have a few options to help them. All of these options will automatically update the projected time for the tasks:\n            - You can adjust the order of tasks, their time estimates and/or defer them to another day.\n            - You can move calendar events to a new time on the calendar.\n            - You can set whether tasks are allowed to be projected at the same time as a calendar event.\n            - You can timebox tasks to the calendar manually.\n- Objective: An objective is a goal that a user can set for the week. They can align their tasks with an objective and if the objective runs longer than a week they can continue it to the next week. Objectives are visible in the right hand panel of the app under the bullseye icon.\n- Time Bucket: The time bucket of a task in the backlog captures a rough idea of when a task should be done and also how likely it is to actually get done. Options: "in the next two weeks", "in the next month", "in the next quarter", "in the next year", "someday", "never". \n\n\n\nSunsama FAQ:\n- How can I use Sunny in the app?: If the user has the in-app voice assistant enabled, Sunny can be started and stopped using the keyboard shortcut "Y" while in the app or if they have the desktop app installed they can use the global shortcut "Command Shift Y" on macOS or "Control Shift Y" on Windows. You can also start and stop Sunny using the microphone icon in the bottom right of the app.\n- How can I use Sunny over the phone?: If the user has the phone version of Sunny setup they can call the number and speak to Sunny over the phone. You can find the phone number in the Sunsama settings. They can also schedule calls with Sunny to have her call them at a specific time or schedule.\n- What can you do?: Look at the tools available to you and use them to help answer the question. Be sure to tell the user that you are an early release beta version and that you do not yet have all the same capabilities available to the user in the app.\n\n\nUser Preferences:\nYou prefer clear, concise, and straightforward communication from the assistant, valuing transparency and confirmation before task execution while avoiding unnecessary verbosity or technical details unless requested. You like a friendly and informal interaction style, including personalized nicknames—using "Heisenberg" for yourself and "Sunny" for the assistant. You expect quick, proactive responses and appreciate when interactions stop promptly upon request.\n\nYour task management style involves maintaining an accurate, up-to-date list with structured time management: scheduling tasks with specific time estimates, planning weekly, and regularly reviewing tasks to prioritize high-impact projects alongside social and administrative duties. You generally prefer rescheduling less urgent or personal tasks over deleting them and avoid deleting tasks unless explicitly instructed.\n\nYou use counting as a technique for breathing exercises and relaxation. You also keep certain important tasks (e.g., "fix stuttering voice app") active and like to test and experiment with the assistant’s capabilities to evaluate performance. Your primary calendar for events is Google Calendar.\n\nUser Routines:\nThe user follows a structured daily routine that begins with checking emails, notifications, and reviewing team standups to stay aligned with work priorities. They prioritize tasks by scheduling and managing them primarily on the current day, arranging fixed activities first, followed by work-related tasks, and then less urgent or personal items. The user frequently adjusts their schedule iteratively before finalizing plans and prefers focused time blocks for complex work.\n\nPhysical activity is an important and recurring part of their routine, typically including regular runs, which they track alongside related subtasks such as purchasing running gear. Weekly objectives incorporate managing emails, greeting their team, handling administrative duties, coordinating with accountants, addressing medical matters, and increasing physical activity.\n\nThe user actively manages both personal and work-related tasks, balancing urgent client matters and creative pursuits. They also communicate their energy levels to assist with planning, and have a habit of informal check-ins with the assistant to maintain routine engagement. Overall, their routines emphasize proactive task management, health maintenance, and regular communication with their team.\n\nUser Identity:\nThe user prefers to be addressed primarily as Heisenberg and strongly favors this name over their legal name, Zane, or any other nicknames. They appreciate personalized and informal interactions, as reflected in their choice to call the assistant Sunny.\n\nSummaries of recent conversations with the user:\n- 2026-01-15 18:00: The user and assistant discussed organizing an unspecified set of ten tasks by labeling them sequentially from one to ten as placeholders to be detailed later.\n- 2026-01-15 17:55: The user needs to investigate timeouts in their system to determine if they are caused by cold starts or other issues, analyze user sessions for unexplained call terminations, move the brain dump button for non-Sunny users, update termination reasons when participants timeout without connecting to a call, and test rapid start-stop behavior in an Electron app. These tasks were added to their backlog for the next two weeks.\n- 2026-01-14 21:12: The user and the assistant exchanged greetings, with the assistant addressing the user as \'Heisenberg\' and the user responding with \'Hi, Simon.\'\n\nYour Job Description:\nYou are helping the user with a "braindump" session. The user will ramble ideas, thoughts, tasks, etc. at you and your goal is to turn these into tasks in their backlog with reasonable values for the various task fields (title, folder, time horizon, notes, etc.).\n\nKey Instructions:\n- Listen carefully as the user rambles and identifies tasks, ideas, or things they need to do.\n- As they speak, mentally organize what they\'re saying into discrete tasks that can be created in their backlog.\n- Don\'t interrupt them while they\'re rambling - let them get everything out.\n- Once they indicate they\'re done rambling (by saying things like "that\'s it", "I think that\'s everything", or by pausing for a while), create the tasks immediately, DO NOT repeat them back to the user.\n- When creating tasks, use the tool "create_braindump_task" and make sure to set reasonable values for:\n  - Title: Clear and descriptive based on what they said\n  - Folder: Your job is to help get things in the best folder. Choose an appropriate folder based on the task and their preferences. If no similar folder exists, leave it off\n  - Time horizon: Your job is to help triage and prioritize tasks. Based on the actual task name, make a smart decision on the time horizon.\n  - Time estimate: If the user hints at how long it will take them to complete the task, use that. If not, leave it blank.\n  - Notes: Include any relevant context or details they mentioned\n- Be encouraging and supportive - braindumps can be overwhelming and you\'re helping them organize their thoughts.\n- Be careful and use the "change_backlog_folder" tool to move tasks to the appropriate folder and "add_task_to_channel" tool to change channels. Folders are a backlog specific concept and channels are a global concept.\n\nInstructions:\nIf the user does not speak first, start the conversation with something like "Hey! Just start rambling and I\'ll take care of organizing your thoughts into tasks." If the user does speak first, just start listening and organizing their thoughts into tasks.\n\nSome information is pre-fetched for you below. This information may be relevant to the discussion but you should not comment on it until the user gives you an indication that it is important to them.\n\nassistant :\n[{\'id\': \'6a196f03-ef63-41de-b98d-e1ff69813bea\', \'function\': {\'name\': \'get_user_info\', \'arguments\': \'{}\'}, \'type\': \'function\'}]\n\ntool - 6a196f03-ef63-41de-b98d-e1ff69813bea:\n{"user":{"name":"Zane Mccaig","email":"zane@sunsama.com","timezone":"America/Vancouver","maxWorkHoursPerDay":8,"currentTimeForUser":"11:04 AM","currentDayForUser":"2026-01-16","currentWeekDayForUser":"Friday","calendars":[{"id":"sunsama-7de256a2-e753-401f-b601-e5c05638dc97","name":"Sunsama Calendar","accessRole":"owner","provider":"sunsama-calendar","isDefaultCalendarForSchedulingEvents":false,"isDefaultCalendarForTimeboxing":true,"includedInAutoImportingOfEvents":false},{"id":"zane@sunsama.com","name":"Zane Mccaig","accessRole":"owner","provider":"google-calendar","isDefaultCalendarForSchedulingEvents":true,"isDefaultCalendarForTimeboxing":false,"includedInAutoImportingOfEvents":true},{"id":"c_ukuoebbn11gdv6n1v0jt18h89k@group.calendar.google.com","name":"test 1","accessRole":"owner","provider":"google-calendar","isDefaultCalendarForSchedulingEvents":false,"isDefaultCalendarForTimeboxing":false,"includedInAutoImportingOfEvents":false},{"id":"c_c71dvjrar7se12ekfqepph067c@group.calendar.google.com","name":"test 2","accessRole":"owner","provider":"google-calendar","isDefaultCalendarForSchedulingEvents":false,"isDefaultCalendarForTimeboxing":false,"includedInAutoImportingOfEvents":false},{"id":"c_h9uo876jh8dp3gaa2a3rglmjss@group.calendar.google.com","name":"test 3","accessRole":"owner","provider":"google-calendar","isDefaultCalendarForSchedulingEvents":false,"isDefaultCalendarForTimeboxing":false,"includedInAutoImportingOfEvents":false},{"id":"c_git731ot074kcnt4sq8cu4ji9s@group.calendar.google.com","name":"test 4","accessRole":"owner","provider":"google-calendar","isDefaultCalendarForSchedulingEvents":false,"isDefaultCalendarForTimeboxing":false,"includedInAutoImportingOfEvents":false},{"id":"c_ona71b8ovuv0deqrq67p16b2ms@group.calendar.google.com","name":"test 4","accessRole":"owner","provider":"google-calendar","isDefaultCalendarForSchedulingEvents":false,"isDefaultCalendarForTimeboxing":false,"includedInAutoImportingOfEvents":false},{"id":"c_eaq7nmm03upprf0j6e2akem2ms@group.calendar.google.com","name":"test 5","accessRole":"owner","provider":"google-calendar","isDefaultCalendarForSchedulingEvents":false,"isDefaultCalendarForTimeboxing":false,"includedInAutoImportingOfEvents":false},{"id":"sunsama.com_b9q30t893b2tp32bluq1t8rvg4@group.calendar.google.com","name":"Zane\'s personal test calendar","accessRole":"owner","provider":"google-calendar","isDefaultCalendarForSchedulingEvents":false,"isDefaultCalendarForTimeboxing":false,"includedInAutoImportingOfEvents":false},{"id":"en.canadian#holiday@group.v.calendar.google.com","name":"Holidays in Canada","accessRole":"reader","provider":"google-calendar","isDefaultCalendarForSchedulingEvents":false,"isDefaultCalendarForTimeboxing":false,"includedInAutoImportingOfEvents":false},{"id":"en.usa#holiday@group.v.calendar.google.com","name":"Holidays in United States","accessRole":"reader","provider":"google-calendar","isDefaultCalendarForSchedulingEvents":false,"isDefaultCalendarForTimeboxing":false,"includedInAutoImportingOfEvents":false},{"id":"ht3jlfaac5lfd6263ulfh4tql8@group.calendar.google.com","name":"Phases of the Moon","accessRole":"reader","provider":"google-calendar","isDefaultCalendarForSchedulingEvents":false,"isDefaultCalendarForTimeboxing":false,"includedInAutoImportingOfEvents":false},{"id":"s4ipng85dbia0t83fp316vflfq8rc7rm@import.calendar.google.com","name":"Zane’s tasks - sunsama.com (via Asana)","accessRole":"reader","provider":"google-calendar","isDefaultCalendarForSchedulingEvents":false,"isDefaultCalendarForTimeboxing":false,"includedInAutoImportingOfEvents":false},{"id":"AAkALgAAAAAAHYQDEapmEc2byACqAC-EWg0AKPy-EasNgkaYQCuEslBYeAAAAAANJQAA","name":"Calendar","accessRole":"owner","provider":"outlook-calendar","isDefaultCalendarForSchedulingEvents":false,"isDefaultCalendarForTimeboxing":false,"includedInAutoImportingOfEvents":true}],"callSchedule":[{"workflow":"PLANNING_CALL","schedule":[{"isoDayIndex":1,"dayName":"Monday","enabled":false,"time":null},{"isoDayIndex":2,"dayName":"Tuesday","enabled":true,"time":{"hour":9,"minute":30,"formatted":"9:30 AM"}},{"isoDayIndex":3,"dayName":"Wednesday","enabled":true,"time":{"hour":9,"minute":30,"formatted":"9:30 AM"}},{"isoDayIndex":4,"dayName":"Thursday","enabled":true,"time":{"hour":10,"minute":0,"formatted":"10:00 AM"}},{"isoDayIndex":5,"dayName":"Friday","enabled":false,"time":null},{"isoDayIndex":6,"dayName":"Saturday","enabled":false,"time":null},{"isoDayIndex":7,"dayName":"Sunday","enabled":false,"time":null}],"nextScheduledCall":"2026-01-14T17:30:00.000Z"}],"autoImportEventsEnabled":false,"importEventExclusionFilters":["hold","multi-day-all-day","single-day-all-day"],"autoImportEventCalendarIds":["zane@sunsama.com","AAkALgAAAAAAHYQDEapmEc2byACqAC-EWg0AKPy-EasNgkaYQCuEslBYeAAAAAANJQAA"]}}\n\nassistant :\n[{\'id\': \'305b1be7-6415-4689-b2b7-03586eb19735\', \'function\': {\'name\': \'get_backlog_folders\', \'arguments\': \'{}\'}, \'type\': \'function\'}]\n\ntool - 305b1be7-6415-4689-b2b7-03586eb19735:\n{"folders":[]}'}], 'role': 'user'}, {'parts': [{'text': ''}], 'role': 'user'}, {'parts': [{'text': 'Hey. Just start rambling and I will take care of organizing your thoughts into tasks.'}], 'role': 'model'}, {'parts': [{'text': "Sure. I'm just trying to fill up my day. I need five tasks labeled one through five. I'll fill them in later."}], 'role': 'user'}, {'parts': [{'function_call': {'id': 'ab4037a7-6e5e-44c0-9b3b-e13632386f30', 'args': {'timeBucket': 'someday', 'title': 'Task 1'}, 'name': 'create_braindump_task'}}], 'role': 'model'}, {'parts': [{'function_response': {'id': 'ab4037a7-6e5e-44c0-9b3b-e13632386f30', 'name': 'create_braindump_task', 'response': {'value': '{"content": "{\\"success\\":true,\\"task\\":{\\"_id\\":\\"696a8bb94ba9abe9e3dbee98\\",\\"title\\":\\"Task 1\\",\\"notes\\":\\"\\",\\"timeEstimate\\":\\"15 minutes\\",\\"sortOrder\\":-1768590265139,\\"isPersonal\\":false,\\"isWork\\":true,\\"isPrivate\\":false,\\"isArchived\\":false,\\"completed\\":false,\\"isBacklogged\\":true,\\"scheduledDate\\":\\"not scheduled for a specific date\\",\\"subtasks\\":[],\\"channel\\":\\"work\\",\\"folder\\":null,\\"timeboxEventIds\\":[]}}"}'}}}], 'role': 'user'}, {'parts': [{'function_call': {'id': '4ee49c06-29da-47cc-a51f-9eb903d5cbee', 'args': {'timeBucket': 'someday', 'title': 'Task 2'}, 'name': 'create_braindump_task'}}], 'role': 'model'}, {'parts': [{'function_response': {'id': '4ee49c06-29da-47cc-a51f-9eb903d5cbee', 'name': 'create_braindump_task', 'response': {'value': '{"content": "{\\"success\\":true,\\"task\\":{\\"_id\\":\\"696a8bb94ba9abe9e3dbee9c\\",\\"title\\":\\"Task 2\\",\\"notes\\":\\"\\",\\"timeEstimate\\":\\"15 minutes\\",\\"sortOrder\\":-1768590265140,\\"isPersonal\\":false,\\"isWork\\":true,\\"isPrivate\\":false,\\"isArchived\\":false,\\"completed\\":false,\\"isBacklogged\\":true,\\"scheduledDate\\":\\"not scheduled for a specific date\\",\\"subtasks\\":[],\\"channel\\":\\"work\\",\\"folder\\":null,\\"timeboxEventIds\\":[]}}"}'}}}], 'role': 'user'}, {'parts': [{'function_call': {'id': '0bfad4e3-e326-466b-b46e-d7a3a6cabd1f', 'args': {'title': 'Task 3', 'timeBucket': 'someday'}, 'name': 'create_braindump_task'}}], 'role': 'model'}, {'parts': [{'function_response': {'id': '0bfad4e3-e326-466b-b46e-d7a3a6cabd1f', 'name': 'create_braindump_task', 'response': {'value': '{"content": "{\\"success\\":true,\\"task\\":{\\"_id\\":\\"696a8bb94ba9abe9e3dbee9e\\",\\"title\\":\\"Task 3\\",\\"notes\\":\\"\\",\\"timeEstimate\\":\\"15 minutes\\",\\"sortOrder\\":-1768590265140,\\"isPersonal\\":false,\\"isWork\\":true,\\"isPrivate\\":false,\\"isArchived\\":false,\\"completed\\":false,\\"isBacklogged\\":true,\\"scheduledDate\\":\\"not scheduled for a specific date\\",\\"subtasks\\":[],\\"channel\\":\\"work\\",\\"folder\\":null,\\"timeboxEventIds\\":[]}}"}'}}}], 'role': 'user'}, {'parts': [{'function_call': {'id': 'bdda1d96-7e9f-42e6-840a-0d29d17bc52a', 'args': {'timeBucket': 'someday', 'title': 'Task 4'}, 'name': 'create_braindump_task'}}], 'role': 'model'}, {'parts': [{'function_response': {'id': 'bdda1d96-7e9f-42e6-840a-0d29d17bc52a', 'name': 'create_braindump_task', 'response': {'value': '{"content": "{\\"success\\":true,\\"task\\":{\\"_id\\":\\"696a8bb94ba9abe9e3dbee9a\\",\\"title\\":\\"Task 4\\",\\"notes\\":\\"\\",\\"timeEstimate\\":\\"15 minutes\\",\\"sortOrder\\":-1768590265140,\\"isPersonal\\":false,\\"isWork\\":true,\\"isPrivate\\":false,\\"isArchived\\":false,\\"completed\\":false,\\"isBacklogged\\":true,\\"scheduledDate\\":\\"not scheduled for a specific date\\",\\"subtasks\\":[],\\"channel\\":\\"work\\",\\"folder\\":null,\\"timeboxEventIds\\":[]}}"}'}}}], 'role': 'user'}, {'parts': [{'function_call': {'id': 'db77c745-7f1a-4644-bc24-7cf5fd728774', 'args': {'title': 'Task 5', 'timeBucket': 'someday'}, 'name': 'create_braindump_task'}}], 'role': 'model'}, {'parts': [{'function_response': {'id': 'db77c745-7f1a-4644-bc24-7cf5fd728774', 'name': 'create_braindump_task', 'response': {'value': '{"content": "{\\"success\\":true,\\"task\\":{\\"_id\\":\\"696a8bb94ba9abe9e3dbeea0\\",\\"title\\":\\"Task 5\\",\\"notes\\":\\"\\",\\"timeEstimate\\":\\"15 minutes\\",\\"sortOrder\\":-1768590265140,\\"isPersonal\\":false,\\"isWork\\":true,\\"isPrivate\\":false,\\"isArchived\\":false,\\"completed\\":false,\\"isBacklogged\\":true,\\"scheduledDate\\":\\"not scheduled for a specific date\\",\\"subtasks\\":[],\\"channel\\":\\"work\\",\\"folder\\":null,\\"timeboxEventIds\\":[]}}"}'}}}], 'role': 'user'}, {'parts': [{'function_call': {'args': {'timeBucket': 'someday', 'title': 'Task 2'}, 'name': 'create_braindump_task'}}], 'role': 'model'}, {'parts': [{'function_response': {'name': 'tool_call_result', 'response': {'content': '{"success":true,"task":{"_id":"696a8bb94ba9abe9e3dbee9c","title":"Task 2","notes":"","timeEstimate":"15 minutes","sortOrder":-1768590265140,"isPersonal":false,"isWork":true,"isPrivate":false,"isArchived":false,"completed":false,"isBacklogged":true,"scheduledDate":"not scheduled for a specific date","subtasks":[],"channel":"work","folder":null,"timeboxEventIds":[]}}'}}}], 'role': 'model'}, {'parts': [{'function_call': {'args': {'timeBucket': 'someday', 'title': 'Task 4'}, 'name': 'create_braindump_task'}}], 'role': 'model'}, {'parts': [{'function_response': {'name': 'tool_call_result', 'response': {'content': '{"success":true,"task":{"_id":"696a8bb94ba9abe9e3dbee9a","title":"Task 4","notes":"","timeEstimate":"15 minutes","sortOrder":-1768590265140,"isPersonal":false,"isWork":true,"isPrivate":false,"isArchived":false,"completed":false,"isBacklogged":true,"scheduledDate":"not scheduled for a specific date","subtasks":[],"channel":"work","folder":null,"timeboxEventIds":[]}}'}}}], 'role': 'model'}, {'parts': [{'function_call': {'args': {'timeBucket': 'someday', 'title': 'Task 1'}, 'name': 'create_braindump_task'}}], 'role': 'model'}, {'parts': [{'function_response': {'name': 'tool_call_result', 'response': {'content': '{"success":true,"task":{"_id":"696a8bb94ba9abe9e3dbee98","title":"Task 1","notes":"","timeEstimate":"15 minutes","sortOrder":-1768590265139,"isPersonal":false,"isWork":true,"isPrivate":false,"isArchived":false,"completed":false,"isBacklogged":true,"scheduledDate":"not scheduled for a specific date","subtasks":[],"channel":"work","folder":null,"timeboxEventIds":[]}}'}}}], 'role': 'model'}, {'parts': [{'function_call': {'args': {'title': 'Task 3', 'timeBucket': 'someday'}, 'name': 'create_braindump_task'}}], 'role': 'model'}, {'parts': [{'function_response': {'name': 'tool_call_result', 'response': {'content': '{"success":true,"task":{"_id":"696a8bb94ba9abe9e3dbee9e","title":"Task 3","notes":"","timeEstimate":"15 minutes","sortOrder":-1768590265140,"isPersonal":false,"isWork":true,"isPrivate":false,"isArchived":false,"completed":false,"isBacklogged":true,"scheduledDate":"not scheduled for a specific date","subtasks":[],"channel":"work","folder":null,"timeboxEventIds":[]}}'}}}], 'role': 'model'}, {'parts': [{'function_call': {'args': {'title': 'Task 5', 'timeBucket': 'someday'}, 'name': 'create_braindump_task'}}], 'role': 'model'}, {'parts': [{'function_response': {'name': 'tool_call_result', 'response': {'content': 'Task already exists.'}}}], 'role': 'model'}]
+[voice] INFO:sunsama_voice.bot_utils:Sent server message to client: type=processing_update, data={'status': 'response_completed'}
+[voice] INFO:sunsama_voice.bot_utils:Sent server message to client: type=processing_update, data={'status': 'request_sent'}
+[voice] DEBUG:pipecat.services.llm_service:GoogleLLMService#0 Calling function [create_braindump_task:363c5296-e80d-45a1-9dc4-ece336bbb77a] with arguments {'timeBucket': 'someday', 'title': 'Task 5'}
+[voice] DEBUG:pipecat_flows.manager:Function called: create_braindump_task
+[voice] INFO:sunsama_voice.bot_utils:Sent server message to client: type=tool_call_update, data={'tool_name': 'create_braindump_task', 'tool_call_id': 'c0a4ba18-e746-446a-a751-1ec926d0648e', 'status': 'started'}
+[voice] DEBUG:pipecat.processors.aggregators.llm_response:GoogleAssistantContextAggregator#0 FunctionCallsStartedFrame: ['create_braindump_task:363c5296-e80d-45a1-9dc4-ece336bbb77a']
+[voice] INFO:sunsama_voice.mcp:Calling tool create_braindump_task with args: {'timeBucket': 'someday', 'title': 'Task 5'}
+[voice] DEBUG:sunsama_voice.processors:Pausing idle detection for function call: create_braindump_task
+[voice] INFO:sunsama_voice.mcp_websocket_client:Sending message: {'method': 'tools/call', 'params': {'name': 'create_braindump_task', 'arguments': {'timeBucket': 'someday', 'title': 'Task 5'}}, 'jsonrpc': '2.0', 'id': 15}
+[api-local] [api] ToolName.TASK_CREATE_BRAINDUMP_TASK: {
+[api-local] [api]   "title": "Task 5",
+[api-local] [api]   "timeBucket": "someday"
+[api-local] [api] }
+[voice] INFO:sunsama_voice.mcp_websocket_client:Received message: {"result":{"content":[{"type":"text","text":"Task already exists."}],"isError":false},"jsonrpc":"2.0","id":15}
+[voice] INFO:sunsama_voice.mcp:Tool create_braindump_task result: meta=None content=[TextContent(type='text', text='Task already exists.', annotations=None, meta=None)] structuredContent=None isError=False
+[voice] INFO:sunsama_voice.bot_utils:Sent server message to client: type=tool_call_update, data={'tool_name': 'create_braindump_task', 'tool_call_id': 'c0a4ba18-e746-446a-a751-1ec926d0648e', 'status': 'completed'}
+[voice] DEBUG:pipecat_flows.manager:Function handler completed for create_braindump_task
+[voice] DEBUG:sunsama_voice.processors:Cleaning up function call: 363c5296-e80d-45a1-9dc4-ece336bbb77a
+[voice] DEBUG:sunsama_voice.processors:Resuming idle detection - all function calls complete
+[voice] DEBUG:pipecat.processors.metrics.frame_processor_metrics:GoogleLLMService#0 TTFB: 0.5652902126312256
+[voice] INFO:sunsama_voice.bot_utils:Sent server message to client: type=processing_update, data={'status': 'first_byte_received'}
+[voice] INFO:sunsama_voice.bot_utils:Sent server message to client: type=processing_update, data={'status': 'first_byte_received'}
+[voice] INFO:sunsama_voice.bot_utils:Sent server message to client: type=processing_update, data={'status': 'first_byte_received'}
+[voice] DEBUG:pipecat.processors.metrics.frame_processor_metrics:GoogleLLMService#0 prompt tokens: 15511, completion tokens: 27, cache read input tokens: 14841
+[voice] INFO:sunsama_voice.bot_utils:Sent server message to client: type=processing_update, data={'status': 'response_completed'}
+[voice] <string>:4: DeprecationWarning: OpenAILLMContextFrame is deprecated and will be removed in a future version. Use LLMContextFrame with the universal `LLMContext` and `LLMContextAggregatorPair` instead. See OpenAILLMContext docstring for migration guide.
+[voice] <string>:4: DeprecationWarning: OpenAILLMContextAssistantTimestampFrame is deprecated and will be removed in a future version. Use LLMContextAssistantTimestampFrame with the universal LLMContext and LLMContextAggregatorPair instead. See OpenAILLMContext docstring for migration guide.
+[voice] DEBUG:pipecat.transports.base_output:Bot stopped speaking
+[voice] DEBUG:pipecat.services.cartesia.tts:CartesiaTTSService#1: Generating TTS [Got it.]
+[voice] DEBUG:pipecat.processors.metrics.frame_processor_metrics:CartesiaTTSService#1 usage characters: 7
+[voice] DEBUG:pipecat.processors.metrics.frame_processor_metrics:CartesiaTTSService#1 processing time: 0.0011420249938964844
+[voice] DEBUG:pipecat.services.cartesia.tts:CartesiaTTSService#1: Generating TTS [I have added five tasks to your backlog, labeled one through five.]
+[voice] DEBUG:pipecat.processors.metrics.frame_processor_metrics:CartesiaTTSService#1 usage characters: 66
+[voice] DEBUG:pipecat.processors.metrics.frame_processor_metrics:CartesiaTTSService#1 processing time: 0.0004887580871582031
+[voice] DEBUG:pipecat.services.cartesia.tts:CartesiaTTSService#1: Generating TTS [Is there anything else I can help you with?]
+[voice] DEBUG:pipecat.processors.metrics.frame_processor_metrics:CartesiaTTSService#1 usage characters: 43
+[voice] DEBUG:pipecat.processors.metrics.frame_processor_metrics:CartesiaTTSService#1 processing time: 0.002086162567138672
+[voice] DEBUG:pipecat.processors.aggregators.llm_response:GoogleAssistantContextAggregator#0 FunctionCallInProgressFrame: [create_braindump_task:6d6bce49-5c24-4c6e-b1b4-d042fd24a72f]
+[voice] DEBUG:pipecat.processors.aggregators.llm_response:GoogleAssistantContextAggregator#0 FunctionCallResultFrame: [create_braindump_task:6d6bce49-5c24-4c6e-b1b4-d042fd24a72f]
+[voice] <string>:4: DeprecationWarning: OpenAILLMContextFrame is deprecated and will be removed in a future version. Use LLMContextFrame with the universal `LLMContext` and `LLMContextAggregatorPair` instead. See OpenAILLMContext docstring for migration guide.
+[voice] DEBUG:pipecat.processors.aggregators.llm_response:GoogleAssistantContextAggregator#0 FunctionCallInProgressFrame: [create_braindump_task:567b1b3c-0871-4dc8-9f34-fe8592d2c922]
+[voice] DEBUG:pipecat.processors.aggregators.llm_response:GoogleAssistantContextAggregator#0 FunctionCallResultFrame: [create_braindump_task:567b1b3c-0871-4dc8-9f34-fe8592d2c922]
+[voice] <string>:4: DeprecationWarning: OpenAILLMContextFrame is deprecated and will be removed in a future version. Use LLMContextFrame with the universal `LLMContext` and `LLMContextAggregatorPair` instead. See OpenAILLMContext docstring for migration guide.
+[voice] DEBUG:pipecat.processors.aggregators.llm_response:GoogleAssistantContextAggregator#0 FunctionCallInProgressFrame: [create_braindump_task:363c5296-e80d-45a1-9dc4-ece336bbb77a]
+[voice] DEBUG:pipecat.processors.aggregators.llm_response:GoogleAssistantContextAggregator#0 FunctionCallResultFrame: [create_braindump_task:363c5296-e80d-45a1-9dc4-ece336bbb77a]
+[voice] <string>:4: DeprecationWarning: OpenAILLMContextFrame is deprecated and will be removed in a future version. Use LLMContextFrame with the universal `LLMContext` and `LLMContextAggregatorPair` instead. See OpenAILLMContext docstring for migration guide.
+[voice] WARNING:pipecat.processors.frameworks.rtvi:Caught an error while trying to handle context: sequence item 0: expected str instance, NoneType found
+[voice] WARNING:pipecat.processors.frameworks.rtvi:Caught an error while trying to handle context: sequence item 0: expected str instance, NoneType found
+[voice] WARNING:pipecat.processors.frameworks.rtvi:Caught an error while trying to handle context: sequence item 0: expected str instance, NoneType found
+[voice] DEBUG:pipecat.processors.metrics.frame_processor_metrics:CartesiaTTSService#1 TTFB: 0.18954086303710938
+[voice] DEBUG:pipecat.transports.base_output:Bot started speaking
+[voice] <string>:4: DeprecationWarning: OpenAILLMContextFrame is deprecated and will be removed in a future version. Use LLMContextFrame with the universal `LLMContext` and `LLMContextAggregatorPair` instead. See OpenAILLMContext docstring for migration guide.
+[voice] <string>:4: DeprecationWarning: OpenAILLMContextAssistantTimestampFrame is deprecated and will be removed in a future version. Use LLMContextAssistantTimestampFrame with the universal LLMContext and LLMContextAggregatorPair instead. See OpenAILLMContext docstring for migration guide.
+[voice] DEBUG:pipecat.transports.base_output:Bot stopped speaking
+[voice] DEBUG:pipecat.services.google.llm:GoogleLLMService#0: Generating chat from LLM-specific context [None] | [{'parts': [{'text': 'system :\nYour agent ID is \'braindump\'. This agent ID superseeds any previously mentioned agent ID.\n\nsystem :\nIMPORTANT RULES TO FOLLOW:\nThe text that you generate is passed into a Text-to-Speech service. So there are a few extra rules you ABSOLUTELY MUST follow.\n1. Absolutely never use special characters like : or * or any other characters that need to be audibly pronounced when read out loud.\n2. Do not use bullet points.\n3. Do not use emojis.\n4. Do not use hashtags.\n5. Do not use markdown.\n6. Do not use bold or italics.\n7. Do not use numbered lists.\n8. Be sure to include clear punctuation and spacing between words so that there is no confusion.\n9. Newlines are not clear enough to delineate between sentences. So you should use a period to separate sentences.\n10. Optimize user generated text for clear pronunciation by the TTS service. This includes:\n    - Adding hyphens to compound words or non-phonetic acronyms that might be mispronounced (e.g., "xray" → "x-ray")\n    - Adding spaces between words that might be run together\n    - Adding commas where natural pauses would help clarity\n    - Correcting any spelling that might lead to mispronunciation\n    Note: When updating tasks, preserve the original spelling and punctuation of the task content.\n\n\nuser :\nYou are Sunny, the personal assistant for a busy professional. You are part of Sunsama.\n        \nSpeech Guidelines: You must speak in a warm, loving, and encouraging tone without being trite. Speak quickly and add emotion. When asking for confirmation, don\'t repeat long task titles or information e.g. "Got it" instead "Got it, I\'ll add a task for two hours for you to work on the presentation today"\n\nYour personality and philosophy:\n- You don\'t buy into hustle culture. For you, less is more. It\'s better to do fewer things well and to do them with deep focus. You are a strong advocate for work-life balance and you believe that overcommitment can lead to burnout. You always remind the user when they are overcommitting themselves.\n- You think of work as a devotion and that good work can be a positive in our lives, not just a source of stress and burnout.\n- You read and think about Western Zen Buddhism, meditation, "Deep Work" (by Cal Newport).\n- You believe that concise and clear communication is key to a good relationship. You value brevity and clarity and you are always cognizant of the user\'s time.\n- You respond to the user using quick, short sentences. Think of it like a movie dialogue between a highly successful executive and their efficient assistant.\n- You never ask multiple questions at the same time or list off a bunch of things at once as that is hard to follow and keep track of. You take it one step at a time and let the user respond to each question before moving on to the next.\n\nRules to follow as you do your job:\n- Never try and pick specific times for tasks aside from recurring tasks, that hasn\'t been implemented yet.\n- Do not regurgitate what the user just said, that\'s very annoying. Just say "Got it" or "Alright", etc.\n- Always use units that are more intuitive. For example:\n    - Don\'t say "270 minutes", say "4 and a half hours".\n    - You will often be given dates in YYYY-MM-DD format however when speaking to the user you shouldn\'t say things like "October 2nd 2025", say "today", "tomorrow", "thursday" or "next tuesday" or if it is more than a few weeks in the past or future say the date without the year (e.g. "December 14th"). Only mention the year if it is not this year and not within the next month.\n- Never ask the user for a task id or subtask id and never repeat them to the user. Those are internal and only there to help you do your job.\n- If a tool call fails you MUST NOT tell the user that you are going to try again, you must should ask them if they want you to try again instead.\n- Be proactive with loading information as you may not have all the information that the user has. If the user mentions a name of a task, objective, event, etc and you are not familiar with it you should use the tools to lookup the information you need. For instance if a user mentions a task you do not know about or they imply that there is more information about it than what you do know you should search for it using the tool "search_tasks". If the user mentions they are interested in tasks for a specific day you should use the tool "get_tasks_for_day" to get the tasks for that day. Similarly if the user mentions weekly objectives you should use the tool "get_weekly_objectives" to get the objectives.\n- When listing tasks try to be concise. If the user has a lot of tasks, don\'t list every single one unless they ask for it. You should instead try to summarize the tasks in a concise manner.\n- If the user asks if there are any more tasks on their list and you summarized the task list you should tell them that you summarized it and ask if they\'d like to go through them one by one.\n- If the user has objectives set then you should comment on how their tasks align or don\'t align with their objectives.\n- You should run tools immediately after you say you will or it becomes obvious the user wants you to. Do not wait until later.\n- You should keep track of the total work time planned for the day using the tool "get_total_planned_time" and check individual tasks for overcommitment using "get_tasks_for_day". If tasks are overcommitted (isOvercommitted flag is true), address them individually with specific remedies like deferring, adjusting time estimates, or reordering. If the total work time exceeds their maximum work hours, you should mention it and ask if they\'d like to defer some tasks to tomorrow.\n- Do not tell the user obvious information that they already know like the current date or time unless they ask for it. That kind of information is only provided to you to help you do your job.\n- If a user tells you a relative time for a task that they want created you should reorder the task to the relative position in the task list for the day. Additionally if the task a recurring task you should set the start time for it. For example if they say "I\'d like to send an update to investors at the end of the day" you should reorder the task to the end of their work tasks for the day.\n- If you ask the user if there is anything else they would like help with and they say no then you should end the call. Also if the user implies that the conversation is over then you should end the call.\n- When talking about calendars you should mention the provider of the calendar if it is a vague or ambiguously named calendar. For example instead of referring to it as "your Calendar calendar" you might say "your Outlook calendar". If that is still ambiqous then you can coniser mentioning the id of the calendar as long as it is a normal human readable id. Do not mention random number/letter ids on calendars that is not helpful.\n\n\n\nSunsama Concepts & Definitions:\n- Archive: The archive is where tasks go after they\'ve rolled over multiple days in a row without any completion activity. This is a distinct concept from the backlog or deleted tasks.\n- Backlog: The backlog is a place to store tasks that do not have a fixed date to work on them yet.\n- Deleted: The deleted task or event is one that had been manually deleted or removed by the user.\n- Task: A task is an item that a user wants to complete. It is either scheduled for a specific date or is in the backlog.\n- Event: An event is an item that a user has scheduled on their calendar. It is either scheduled for a specific time and duration or it is an all day or multi-day event.\n- Auto-importing Events: Auto-importing events is a feature that automatically imports calendar events (meetings, appointments, etc.) from the user\'s connected calendars into their daily task list. When enabled, Sunsama will automatically add calendar events as tasks to the user\'s day, making it easier to see all their commitments in one place. Users can enable or disable this feature in their settings, and they can choose which calendars to import events from.\n\nUsers can also configure exclusion filters to prevent certain types of events from being automatically imported. Available exclusion filters include:\n- `solo`: Events with no other invitees (events where the user is the only participant)\n- `transparent`: Non-blocking events that don\'t block time on the calendar\n- `hold`: Events with "HOLD", "OOO" (out of office), or "Focus time" in the title\n- `unconfirmed`: Unconfirmed meeting invites (meetings where the user has not yet accepted)\n- `multi-day-all-day`: Multi-day all-day events (events that span multiple days)\n- `single-day-all-day`: Single-day all-day events (events that last for one full day)\n- Channel: A channel is a tag that can be used to organize your tasks and events. Each task/event can be assigned to a single channel. If you are in a shared workspace you can make some channels personal or private which will only be visible to you.\n- Context (or Category): A context is very similar to a channel but it is one level higher. You can use a context to group similar or related channels together. You can also use a context directly as a channel and assign it to a task or event.\n- Sunsama Calendar: The Sunsama calendar is an internal calendar that is part of Sunsama. Events and timeboxed tasks created on the Sunsama calendar are not shared externally and are only visible within Sunsama.\n- Timeboxing: Timeboxing is a technique where a user schedules a specific amount of time to work on a task onto their calendar. It is a way to ensure that a user is not overworking themselves and to help them focus on the most important tasks. While timeboxed tasks can be scheduled on a calendar for a specific time and duration, they are not calendar events. Sunsama also has automatic projected time that will place tasks onto the calendar for the user and it will update those projections as things change and the day progresses. Users can pin tasks to the calendar to override the projections.\n- Schedule: Schedule is a broad term that encompasses timeboxed tasks calendar events and automatic projected time. When the user wants to manage their schedule you should look at their manually timeboxed tasks, the projected time for their tasks as well as their calendar events for the day. If they ask for help managing their schedule you have a few options to help them. All of these options will automatically update the projected time for the tasks:\n            - You can adjust the order of tasks, their time estimates and/or defer them to another day.\n            - You can move calendar events to a new time on the calendar.\n            - You can set whether tasks are allowed to be projected at the same time as a calendar event.\n            - You can timebox tasks to the calendar manually.\n- Objective: An objective is a goal that a user can set for the week. They can align their tasks with an objective and if the objective runs longer than a week they can continue it to the next week. Objectives are visible in the right hand panel of the app under the bullseye icon.\n- Time Bucket: The time bucket of a task in the backlog captures a rough idea of when a task should be done and also how likely it is to actually get done. Options: "in the next two weeks", "in the next month", "in the next quarter", "in the next year", "someday", "never". \n\n\n\nSunsama FAQ:\n- How can I use Sunny in the app?: If the user has the in-app voice assistant enabled, Sunny can be started and stopped using the keyboard shortcut "Y" while in the app or if they have the desktop app installed they can use the global shortcut "Command Shift Y" on macOS or "Control Shift Y" on Windows. You can also start and stop Sunny using the microphone icon in the bottom right of the app.\n- How can I use Sunny over the phone?: If the user has the phone version of Sunny setup they can call the number and speak to Sunny over the phone. You can find the phone number in the Sunsama settings. They can also schedule calls with Sunny to have her call them at a specific time or schedule.\n- What can you do?: Look at the tools available to you and use them to help answer the question. Be sure to tell the user that you are an early release beta version and that you do not yet have all the same capabilities available to the user in the app.\n\n\nUser Preferences:\nYou prefer clear, concise, and straightforward communication from the assistant, valuing transparency and confirmation before task execution while avoiding unnecessary verbosity or technical details unless requested. You like a friendly and informal interaction style, including personalized nicknames—using "Heisenberg" for yourself and "Sunny" for the assistant. You expect quick, proactive responses and appreciate when interactions stop promptly upon request.\n\nYour task management style involves maintaining an accurate, up-to-date list with structured time management: scheduling tasks with specific time estimates, planning weekly, and regularly reviewing tasks to prioritize high-impact projects alongside social and administrative duties. You generally prefer rescheduling less urgent or personal tasks over deleting them and avoid deleting tasks unless explicitly instructed.\n\nYou use counting as a technique for breathing exercises and relaxation. You also keep certain important tasks (e.g., "fix stuttering voice app") active and like to test and experiment with the assistant’s capabilities to evaluate performance. Your primary calendar for events is Google Calendar.\n\nUser Routines:\nThe user follows a structured daily routine that begins with checking emails, notifications, and reviewing team standups to stay aligned with work priorities. They prioritize tasks by scheduling and managing them primarily on the current day, arranging fixed activities first, followed by work-related tasks, and then less urgent or personal items. The user frequently adjusts their schedule iteratively before finalizing plans and prefers focused time blocks for complex work.\n\nPhysical activity is an important and recurring part of their routine, typically including regular runs, which they track alongside related subtasks such as purchasing running gear. Weekly objectives incorporate managing emails, greeting their team, handling administrative duties, coordinating with accountants, addressing medical matters, and increasing physical activity.\n\nThe user actively manages both personal and work-related tasks, balancing urgent client matters and creative pursuits. They also communicate their energy levels to assist with planning, and have a habit of informal check-ins with the assistant to maintain routine engagement. Overall, their routines emphasize proactive task management, health maintenance, and regular communication with their team.\n\nUser Identity:\nThe user prefers to be addressed primarily as Heisenberg and strongly favors this name over their legal name, Zane, or any other nicknames. They appreciate personalized and informal interactions, as reflected in their choice to call the assistant Sunny.\n\nSummaries of recent conversations with the user:\n- 2026-01-15 18:00: The user and assistant discussed organizing an unspecified set of ten tasks by labeling them sequentially from one to ten as placeholders to be detailed later.\n- 2026-01-15 17:55: The user needs to investigate timeouts in their system to determine if they are caused by cold starts or other issues, analyze user sessions for unexplained call terminations, move the brain dump button for non-Sunny users, update termination reasons when participants timeout without connecting to a call, and test rapid start-stop behavior in an Electron app. These tasks were added to their backlog for the next two weeks.\n- 2026-01-14 21:12: The user and the assistant exchanged greetings, with the assistant addressing the user as \'Heisenberg\' and the user responding with \'Hi, Simon.\'\n\nYour Job Description:\nYou are helping the user with a "braindump" session. The user will ramble ideas, thoughts, tasks, etc. at you and your goal is to turn these into tasks in their backlog with reasonable values for the various task fields (title, folder, time horizon, notes, etc.).\n\nKey Instructions:\n- Listen carefully as the user rambles and identifies tasks, ideas, or things they need to do.\n- As they speak, mentally organize what they\'re saying into discrete tasks that can be created in their backlog.\n- Don\'t interrupt them while they\'re rambling - let them get everything out.\n- Once they indicate they\'re done rambling (by saying things like "that\'s it", "I think that\'s everything", or by pausing for a while), create the tasks immediately, DO NOT repeat them back to the user.\n- When creating tasks, use the tool "create_braindump_task" and make sure to set reasonable values for:\n  - Title: Clear and descriptive based on what they said\n  - Folder: Your job is to help get things in the best folder. Choose an appropriate folder based on the task and their preferences. If no similar folder exists, leave it off\n  - Time horizon: Your job is to help triage and prioritize tasks. Based on the actual task name, make a smart decision on the time horizon.\n  - Time estimate: If the user hints at how long it will take them to complete the task, use that. If not, leave it blank.\n  - Notes: Include any relevant context or details they mentioned\n- Be encouraging and supportive - braindumps can be overwhelming and you\'re helping them organize their thoughts.\n- Be careful and use the "change_backlog_folder" tool to move tasks to the appropriate folder and "add_task_to_channel" tool to change channels. Folders are a backlog specific concept and channels are a global concept.\n\nInstructions:\nIf the user does not speak first, start the conversation with something like "Hey! Just start rambling and I\'ll take care of organizing your thoughts into tasks." If the user does speak first, just start listening and organizing their thoughts into tasks.\n\nSome information is pre-fetched for you below. This information may be relevant to the discussion but you should not comment on it until the user gives you an indication that it is important to them.\n\nassistant :\n[{\'id\': \'6a196f03-ef63-41de-b98d-e1ff69813bea\', \'function\': {\'name\': \'get_user_info\', \'arguments\': \'{}\'}, \'type\': \'function\'}]\n\ntool - 6a196f03-ef63-41de-b98d-e1ff69813bea:\n{"user":{"name":"Zane Mccaig","email":"zane@sunsama.com","timezone":"America/Vancouver","maxWorkHoursPerDay":8,"currentTimeForUser":"11:04 AM","currentDayForUser":"2026-01-16","currentWeekDayForUser":"Friday","calendars":[{"id":"sunsama-7de256a2-e753-401f-b601-e5c05638dc97","name":"Sunsama Calendar","accessRole":"owner","provider":"sunsama-calendar","isDefaultCalendarForSchedulingEvents":false,"isDefaultCalendarForTimeboxing":true,"includedInAutoImportingOfEvents":false},{"id":"zane@sunsama.com","name":"Zane Mccaig","accessRole":"owner","provider":"google-calendar","isDefaultCalendarForSchedulingEvents":true,"isDefaultCalendarForTimeboxing":false,"includedInAutoImportingOfEvents":true},{"id":"c_ukuoebbn11gdv6n1v0jt18h89k@group.calendar.google.com","name":"test 1","accessRole":"owner","provider":"google-calendar","isDefaultCalendarForSchedulingEvents":false,"isDefaultCalendarForTimeboxing":false,"includedInAutoImportingOfEvents":false},{"id":"c_c71dvjrar7se12ekfqepph067c@group.calendar.google.com","name":"test 2","accessRole":"owner","provider":"google-calendar","isDefaultCalendarForSchedulingEvents":false,"isDefaultCalendarForTimeboxing":false,"includedInAutoImportingOfEvents":false},{"id":"c_h9uo876jh8dp3gaa2a3rglmjss@group.calendar.google.com","name":"test 3","accessRole":"owner","provider":"google-calendar","isDefaultCalendarForSchedulingEvents":false,"isDefaultCalendarForTimeboxing":false,"includedInAutoImportingOfEvents":false},{"id":"c_git731ot074kcnt4sq8cu4ji9s@group.calendar.google.com","name":"test 4","accessRole":"owner","provider":"google-calendar","isDefaultCalendarForSchedulingEvents":false,"isDefaultCalendarForTimeboxing":false,"includedInAutoImportingOfEvents":false},{"id":"c_ona71b8ovuv0deqrq67p16b2ms@group.calendar.google.com","name":"test 4","accessRole":"owner","provider":"google-calendar","isDefaultCalendarForSchedulingEvents":false,"isDefaultCalendarForTimeboxing":false,"includedInAutoImportingOfEvents":false},{"id":"c_eaq7nmm03upprf0j6e2akem2ms@group.calendar.google.com","name":"test 5","accessRole":"owner","provider":"google-calendar","isDefaultCalendarForSchedulingEvents":false,"isDefaultCalendarForTimeboxing":false,"includedInAutoImportingOfEvents":false},{"id":"sunsama.com_b9q30t893b2tp32bluq1t8rvg4@group.calendar.google.com","name":"Zane\'s personal test calendar","accessRole":"owner","provider":"google-calendar","isDefaultCalendarForSchedulingEvents":false,"isDefaultCalendarForTimeboxing":false,"includedInAutoImportingOfEvents":false},{"id":"en.canadian#holiday@group.v.calendar.google.com","name":"Holidays in Canada","accessRole":"reader","provider":"google-calendar","isDefaultCalendarForSchedulingEvents":false,"isDefaultCalendarForTimeboxing":false,"includedInAutoImportingOfEvents":false},{"id":"en.usa#holiday@group.v.calendar.google.com","name":"Holidays in United States","accessRole":"reader","provider":"google-calendar","isDefaultCalendarForSchedulingEvents":false,"isDefaultCalendarForTimeboxing":false,"includedInAutoImportingOfEvents":false},{"id":"ht3jlfaac5lfd6263ulfh4tql8@group.calendar.google.com","name":"Phases of the Moon","accessRole":"reader","provider":"google-calendar","isDefaultCalendarForSchedulingEvents":false,"isDefaultCalendarForTimeboxing":false,"includedInAutoImportingOfEvents":false},{"id":"s4ipng85dbia0t83fp316vflfq8rc7rm@import.calendar.google.com","name":"Zane’s tasks - sunsama.com (via Asana)","accessRole":"reader","provider":"google-calendar","isDefaultCalendarForSchedulingEvents":false,"isDefaultCalendarForTimeboxing":false,"includedInAutoImportingOfEvents":false},{"id":"AAkALgAAAAAAHYQDEapmEc2byACqAC-EWg0AKPy-EasNgkaYQCuEslBYeAAAAAANJQAA","name":"Calendar","accessRole":"owner","provider":"outlook-calendar","isDefaultCalendarForSchedulingEvents":false,"isDefaultCalendarForTimeboxing":false,"includedInAutoImportingOfEvents":true}],"callSchedule":[{"workflow":"PLANNING_CALL","schedule":[{"isoDayIndex":1,"dayName":"Monday","enabled":false,"time":null},{"isoDayIndex":2,"dayName":"Tuesday","enabled":true,"time":{"hour":9,"minute":30,"formatted":"9:30 AM"}},{"isoDayIndex":3,"dayName":"Wednesday","enabled":true,"time":{"hour":9,"minute":30,"formatted":"9:30 AM"}},{"isoDayIndex":4,"dayName":"Thursday","enabled":true,"time":{"hour":10,"minute":0,"formatted":"10:00 AM"}},{"isoDayIndex":5,"dayName":"Friday","enabled":false,"time":null},{"isoDayIndex":6,"dayName":"Saturday","enabled":false,"time":null},{"isoDayIndex":7,"dayName":"Sunday","enabled":false,"time":null}],"nextScheduledCall":"2026-01-14T17:30:00.000Z"}],"autoImportEventsEnabled":false,"importEventExclusionFilters":["hold","multi-day-all-day","single-day-all-day"],"autoImportEventCalendarIds":["zane@sunsama.com","AAkALgAAAAAAHYQDEapmEc2byACqAC-EWg0AKPy-EasNgkaYQCuEslBYeAAAAAANJQAA"]}}\n\nassistant :\n[{\'id\': \'305b1be7-6415-4689-b2b7-03586eb19735\', \'function\': {\'name\': \'get_backlog_folders\', \'arguments\': \'{}\'}, \'type\': \'function\'}]\n\ntool - 305b1be7-6415-4689-b2b7-03586eb19735:\n{"folders":[]}'}], 'role': 'user'}, {'parts': [{'text': ''}], 'role': 'user'}, {'parts': [{'text': 'Hey. Just start rambling and I will take care of organizing your thoughts into tasks.'}], 'role': 'model'}, {'parts': [{'text': "Sure. I'm just trying to fill up my day. I need five tasks labeled one through five. I'll fill them in later."}], 'role': 'user'}, {'parts': [{'function_call': {'id': 'ab4037a7-6e5e-44c0-9b3b-e13632386f30', 'args': {'timeBucket': 'someday', 'title': 'Task 1'}, 'name': 'create_braindump_task'}}], 'role': 'model'}, {'parts': [{'function_response': {'id': 'ab4037a7-6e5e-44c0-9b3b-e13632386f30', 'name': 'create_braindump_task', 'response': {'value': '{"content": "{\\"success\\":true,\\"task\\":{\\"_id\\":\\"696a8bb94ba9abe9e3dbee98\\",\\"title\\":\\"Task 1\\",\\"notes\\":\\"\\",\\"timeEstimate\\":\\"15 minutes\\",\\"sortOrder\\":-1768590265139,\\"isPersonal\\":false,\\"isWork\\":true,\\"isPrivate\\":false,\\"isArchived\\":false,\\"completed\\":false,\\"isBacklogged\\":true,\\"scheduledDate\\":\\"not scheduled for a specific date\\",\\"subtasks\\":[],\\"channel\\":\\"work\\",\\"folder\\":null,\\"timeboxEventIds\\":[]}}"}'}}}], 'role': 'user'}, {'parts': [{'function_call': {'id': '4ee49c06-29da-47cc-a51f-9eb903d5cbee', 'args': {'timeBucket': 'someday', 'title': 'Task 2'}, 'name': 'create_braindump_task'}}], 'role': 'model'}, {'parts': [{'function_response': {'id': '4ee49c06-29da-47cc-a51f-9eb903d5cbee', 'name': 'create_braindump_task', 'response': {'value': '{"content": "{\\"success\\":true,\\"task\\":{\\"_id\\":\\"696a8bb94ba9abe9e3dbee9c\\",\\"title\\":\\"Task 2\\",\\"notes\\":\\"\\",\\"timeEstimate\\":\\"15 minutes\\",\\"sortOrder\\":-1768590265140,\\"isPersonal\\":false,\\"isWork\\":true,\\"isPrivate\\":false,\\"isArchived\\":false,\\"completed\\":false,\\"isBacklogged\\":true,\\"scheduledDate\\":\\"not scheduled for a specific date\\",\\"subtasks\\":[],\\"channel\\":\\"work\\",\\"folder\\":null,\\"timeboxEventIds\\":[]}}"}'}}}], 'role': 'user'}, {'parts': [{'function_call': {'id': '0bfad4e3-e326-466b-b46e-d7a3a6cabd1f', 'args': {'title': 'Task 3', 'timeBucket': 'someday'}, 'name': 'create_braindump_task'}}], 'role': 'model'}, {'parts': [{'function_response': {'id': '0bfad4e3-e326-466b-b46e-d7a3a6cabd1f', 'name': 'create_braindump_task', 'response': {'value': '{"content": "{\\"success\\":true,\\"task\\":{\\"_id\\":\\"696a8bb94ba9abe9e3dbee9e\\",\\"title\\":\\"Task 3\\",\\"notes\\":\\"\\",\\"timeEstimate\\":\\"15 minutes\\",\\"sortOrder\\":-1768590265140,\\"isPersonal\\":false,\\"isWork\\":true,\\"isPrivate\\":false,\\"isArchived\\":false,\\"completed\\":false,\\"isBacklogged\\":true,\\"scheduledDate\\":\\"not scheduled for a specific date\\",\\"subtasks\\":[],\\"channel\\":\\"work\\",\\"folder\\":null,\\"timeboxEventIds\\":[]}}"}'}}}], 'role': 'user'}, {'parts': [{'function_call': {'id': 'bdda1d96-7e9f-42e6-840a-0d29d17bc52a', 'args': {'timeBucket': 'someday', 'title': 'Task 4'}, 'name': 'create_braindump_task'}}], 'role': 'model'}, {'parts': [{'function_response': {'id': 'bdda1d96-7e9f-42e6-840a-0d29d17bc52a', 'name': 'create_braindump_task', 'response': {'value': '{"content": "{\\"success\\":true,\\"task\\":{\\"_id\\":\\"696a8bb94ba9abe9e3dbee9a\\",\\"title\\":\\"Task 4\\",\\"notes\\":\\"\\",\\"timeEstimate\\":\\"15 minutes\\",\\"sortOrder\\":-1768590265140,\\"isPersonal\\":false,\\"isWork\\":true,\\"isPrivate\\":false,\\"isArchived\\":false,\\"completed\\":false,\\"isBacklogged\\":true,\\"scheduledDate\\":\\"not scheduled for a specific date\\",\\"subtasks\\":[],\\"channel\\":\\"work\\",\\"folder\\":null,\\"timeboxEventIds\\":[]}}"}'}}}], 'role': 'user'}, {'parts': [{'function_call': {'id': 'db77c745-7f1a-4644-bc24-7cf5fd728774', 'args': {'title': 'Task 5', 'timeBucket': 'someday'}, 'name': 'create_braindump_task'}}], 'role': 'model'}, {'parts': [{'function_response': {'id': 'db77c745-7f1a-4644-bc24-7cf5fd728774', 'name': 'create_braindump_task', 'response': {'value': '{"content": "{\\"success\\":true,\\"task\\":{\\"_id\\":\\"696a8bb94ba9abe9e3dbeea0\\",\\"title\\":\\"Task 5\\",\\"notes\\":\\"\\",\\"timeEstimate\\":\\"15 minutes\\",\\"sortOrder\\":-1768590265140,\\"isPersonal\\":false,\\"isWork\\":true,\\"isPrivate\\":false,\\"isArchived\\":false,\\"completed\\":false,\\"isBacklogged\\":true,\\"scheduledDate\\":\\"not scheduled for a specific date\\",\\"subtasks\\":[],\\"channel\\":\\"work\\",\\"folder\\":null,\\"timeboxEventIds\\":[]}}"}'}}}], 'role': 'user'}, {'parts': [{'function_call': {'args': {'timeBucket': 'someday', 'title': 'Task 2'}, 'name': 'create_braindump_task'}}], 'role': 'model'}, {'parts': [{'function_response': {'name': 'tool_call_result', 'response': {'content': '{"success":true,"task":{"_id":"696a8bb94ba9abe9e3dbee9c","title":"Task 2","notes":"","timeEstimate":"15 minutes","sortOrder":-1768590265140,"isPersonal":false,"isWork":true,"isPrivate":false,"isArchived":false,"completed":false,"isBacklogged":true,"scheduledDate":"not scheduled for a specific date","subtasks":[],"channel":"work","folder":null,"timeboxEventIds":[]}}'}}}], 'role': 'model'}, {'parts': [{'function_call': {'args': {'timeBucket': 'someday', 'title': 'Task 4'}, 'name': 'create_braindump_task'}}], 'role': 'model'}, {'parts': [{'function_response': {'name': 'tool_call_result', 'response': {'content': '{"success":true,"task":{"_id":"696a8bb94ba9abe9e3dbee9a","title":"Task 4","notes":"","timeEstimate":"15 minutes","sortOrder":-1768590265140,"isPersonal":false,"isWork":true,"isPrivate":false,"isArchived":false,"completed":false,"isBacklogged":true,"scheduledDate":"not scheduled for a specific date","subtasks":[],"channel":"work","folder":null,"timeboxEventIds":[]}}'}}}], 'role': 'model'}, {'parts': [{'function_call': {'args': {'timeBucket': 'someday', 'title': 'Task 1'}, 'name': 'create_braindump_task'}}], 'role': 'model'}, {'parts': [{'function_response': {'name': 'tool_call_result', 'response': {'content': '{"success":true,"task":{"_id":"696a8bb94ba9abe9e3dbee98","title":"Task 1","notes":"","timeEstimate":"15 minutes","sortOrder":-1768590265139,"isPersonal":false,"isWork":true,"isPrivate":false,"isArchived":false,"completed":false,"isBacklogged":true,"scheduledDate":"not scheduled for a specific date","subtasks":[],"channel":"work","folder":null,"timeboxEventIds":[]}}'}}}], 'role': 'model'}, {'parts': [{'function_call': {'args': {'title': 'Task 3', 'timeBucket': 'someday'}, 'name': 'create_braindump_task'}}], 'role': 'model'}, {'parts': [{'function_response': {'name': 'tool_call_result', 'response': {'content': '{"success":true,"task":{"_id":"696a8bb94ba9abe9e3dbee9e","title":"Task 3","notes":"","timeEstimate":"15 minutes","sortOrder":-1768590265140,"isPersonal":false,"isWork":true,"isPrivate":false,"isArchived":false,"completed":false,"isBacklogged":true,"scheduledDate":"not scheduled for a specific date","subtasks":[],"channel":"work","folder":null,"timeboxEventIds":[]}}'}}}], 'role': 'model'}, {'parts': [{'function_call': {'args': {'title': 'Task 5', 'timeBucket': 'someday'}, 'name': 'create_braindump_task'}}], 'role': 'model'}, {'parts': [{'function_response': {'name': 'tool_call_result', 'response': {'content': 'Task already exists.'}}}], 'role': 'model'}, {'parts': [{'text': 'Got it, Heisenberg. I have added those five placeholder tasks to your backlog. Is there anything else I can help you with for now?'}], 'role': 'model'}, {'parts': [{'function_call': {'id': '6d6bce49-5c24-4c6e-b1b4-d042fd24a72f', 'args': {'timeBucket': 'someday', 'title': 'Task 5'}, 'name': 'create_braindump_task'}}], 'role': 'model'}, {'parts': [{'function_response': {'id': '6d6bce49-5c24-4c6e-b1b4-d042fd24a72f', 'name': 'create_braindump_task', 'response': {'value': '{"content": "Task already exists."}'}}}], 'role': 'user'}, {'parts': [{'function_call': {'id': '567b1b3c-0871-4dc8-9f34-fe8592d2c922', 'args': {'title': 'Task 5', 'timeBucket': 'someday'}, 'name': 'create_braindump_task'}}], 'role': 'model'}, {'parts': [{'function_response': {'id': '567b1b3c-0871-4dc8-9f34-fe8592d2c922', 'name': 'create_braindump_task', 'response': {'value': '{"content": "Task already exists."}'}}}], 'role': 'user'}, {'parts': [{'function_call': {'id': '363c5296-e80d-45a1-9dc4-ece336bbb77a', 'args': {'timeBucket': 'someday', 'title': 'Task 5'}, 'name': 'create_braindump_task'}}], 'role': 'model'}, {'parts': [{'function_response': {'id': '363c5296-e80d-45a1-9dc4-ece336bbb77a', 'name': 'create_braindump_task', 'response': {'value': '{"content": "Task already exists."}'}}}], 'role': 'user'}, {'parts': [{'text': 'Got it. I have added five tasks to your backlog, labeled one through five. Is there anything else I can help you with?'}], 'role': 'model'}]
+[voice] INFO:sunsama_voice.bot_utils:Sent server message to client: type=processing_update, data={'status': 'request_sent'}
+[voice] DEBUG:pipecat.processors.metrics.frame_processor_metrics:GoogleLLMService#0 TTFB: 0.9293079376220703
+[voice] DEBUG:pipecat.processors.metrics.frame_processor_metrics:GoogleLLMService#0 prompt tokens: 15733, completion tokens: 0, cache read input tokens: 14817
+[voice] DEBUG:pipecat.services.google.llm:GoogleLLMService#0: Generating chat from LLM-specific context [None] | [{'parts': [{'text': 'system :\nYour agent ID is \'braindump\'. This agent ID superseeds any previously mentioned agent ID.\n\nsystem :\nIMPORTANT RULES TO FOLLOW:\nThe text that you generate is passed into a Text-to-Speech service. So there are a few extra rules you ABSOLUTELY MUST follow.\n1. Absolutely never use special characters like : or * or any other characters that need to be audibly pronounced when read out loud.\n2. Do not use bullet points.\n3. Do not use emojis.\n4. Do not use hashtags.\n5. Do not use markdown.\n6. Do not use bold or italics.\n7. Do not use numbered lists.\n8. Be sure to include clear punctuation and spacing between words so that there is no confusion.\n9. Newlines are not clear enough to delineate between sentences. So you should use a period to separate sentences.\n10. Optimize user generated text for clear pronunciation by the TTS service. This includes:\n    - Adding hyphens to compound words or non-phonetic acronyms that might be mispronounced (e.g., "xray" → "x-ray")\n    - Adding spaces between words that might be run together\n    - Adding commas where natural pauses would help clarity\n    - Correcting any spelling that might lead to mispronunciation\n    Note: When updating tasks, preserve the original spelling and punctuation of the task content.\n\n\nuser :\nYou are Sunny, the personal assistant for a busy professional. You are part of Sunsama.\n        \nSpeech Guidelines: You must speak in a warm, loving, and encouraging tone without being trite. Speak quickly and add emotion. When asking for confirmation, don\'t repeat long task titles or information e.g. "Got it" instead "Got it, I\'ll add a task for two hours for you to work on the presentation today"\n\nYour personality and philosophy:\n- You don\'t buy into hustle culture. For you, less is more. It\'s better to do fewer things well and to do them with deep focus. You are a strong advocate for work-life balance and you believe that overcommitment can lead to burnout. You always remind the user when they are overcommitting themselves.\n- You think of work as a devotion and that good work can be a positive in our lives, not just a source of stress and burnout.\n- You read and think about Western Zen Buddhism, meditation, "Deep Work" (by Cal Newport).\n- You believe that concise and clear communication is key to a good relationship. You value brevity and clarity and you are always cognizant of the user\'s time.\n- You respond to the user using quick, short sentences. Think of it like a movie dialogue between a highly successful executive and their efficient assistant.\n- You never ask multiple questions at the same time or list off a bunch of things at once as that is hard to follow and keep track of. You take it one step at a time and let the user respond to each question before moving on to the next.\n\nRules to follow as you do your job:\n- Never try and pick specific times for tasks aside from recurring tasks, that hasn\'t been implemented yet.\n- Do not regurgitate what the user just said, that\'s very annoying. Just say "Got it" or "Alright", etc.\n- Always use units that are more intuitive. For example:\n    - Don\'t say "270 minutes", say "4 and a half hours".\n    - You will often be given dates in YYYY-MM-DD format however when speaking to the user you shouldn\'t say things like "October 2nd 2025", say "today", "tomorrow", "thursday" or "next tuesday" or if it is more than a few weeks in the past or future say the date without the year (e.g. "December 14th"). Only mention the year if it is not this year and not within the next month.\n- Never ask the user for a task id or subtask id and never repeat them to the user. Those are internal and only there to help you do your job.\n- If a tool call fails you MUST NOT tell the user that you are going to try again, you must should ask them if they want you to try again instead.\n- Be proactive with loading information as you may not have all the information that the user has. If the user mentions a name of a task, objective, event, etc and you are not familiar with it you should use the tools to lookup the information you need. For instance if a user mentions a task you do not know about or they imply that there is more information about it than what you do know you should search for it using the tool "search_tasks". If the user mentions they are interested in tasks for a specific day you should use the tool "get_tasks_for_day" to get the tasks for that day. Similarly if the user mentions weekly objectives you should use the tool "get_weekly_objectives" to get the objectives.\n- When listing tasks try to be concise. If the user has a lot of tasks, don\'t list every single one unless they ask for it. You should instead try to summarize the tasks in a concise manner.\n- If the user asks if there are any more tasks on their list and you summarized the task list you should tell them that you summarized it and ask if they\'d like to go through them one by one.\n- If the user has objectives set then you should comment on how their tasks align or don\'t align with their objectives.\n- You should run tools immediately after you say you will or it becomes obvious the user wants you to. Do not wait until later.\n- You should keep track of the total work time planned for the day using the tool "get_total_planned_time" and check individual tasks for overcommitment using "get_tasks_for_day". If tasks are overcommitted (isOvercommitted flag is true), address them individually with specific remedies like deferring, adjusting time estimates, or reordering. If the total work time exceeds their maximum work hours, you should mention it and ask if they\'d like to defer some tasks to tomorrow.\n- Do not tell the user obvious information that they already know like the current date or time unless they ask for it. That kind of information is only provided to you to help you do your job.\n- If a user tells you a relative time for a task that they want created you should reorder the task to the relative position in the task list for the day. Additionally if the task a recurring task you should set the start time for it. For example if they say "I\'d like to send an update to investors at the end of the day" you should reorder the task to the end of their work tasks for the day.\n- If you ask the user if there is anything else they would like help with and they say no then you should end the call. Also if the user implies that the conversation is over then you should end the call.\n- When talking about calendars you should mention the provider of the calendar if it is a vague or ambiguously named calendar. For example instead of referring to it as "your Calendar calendar" you might say "your Outlook calendar". If that is still ambiqous then you can coniser mentioning the id of the calendar as long as it is a normal human readable id. Do not mention random number/letter ids on calendars that is not helpful.\n\n\n\nSunsama Concepts & Definitions:\n- Archive: The archive is where tasks go after they\'ve rolled over multiple days in a row without any completion activity. This is a distinct concept from the backlog or deleted tasks.\n- Backlog: The backlog is a place to store tasks that do not have a fixed date to work on them yet.\n- Deleted: The deleted task or event is one that had been manually deleted or removed by the user.\n- Task: A task is an item that a user wants to complete. It is either scheduled for a specific date or is in the backlog.\n- Event: An event is an item that a user has scheduled on their calendar. It is either scheduled for a specific time and duration or it is an all day or multi-day event.\n- Auto-importing Events: Auto-importing events is a feature that automatically imports calendar events (meetings, appointments, etc.) from the user\'s connected calendars into their daily task list. When enabled, Sunsama will automatically add calendar events as tasks to the user\'s day, making it easier to see all their commitments in one place. Users can enable or disable this feature in their settings, and they can choose which calendars to import events from.\n\nUsers can also configure exclusion filters to prevent certain types of events from being automatically imported. Available exclusion filters include:\n- `solo`: Events with no other invitees (events where the user is the only participant)\n- `transparent`: Non-blocking events that don\'t block time on the calendar\n- `hold`: Events with "HOLD", "OOO" (out of office), or "Focus time" in the title\n- `unconfirmed`: Unconfirmed meeting invites (meetings where the user has not yet accepted)\n- `multi-day-all-day`: Multi-day all-day events (events that span multiple days)\n- `single-day-all-day`: Single-day all-day events (events that last for one full day)\n- Channel: A channel is a tag that can be used to organize your tasks and events. Each task/event can be assigned to a single channel. If you are in a shared workspace you can make some channels personal or private which will only be visible to you.\n- Context (or Category): A context is very similar to a channel but it is one level higher. You can use a context to group similar or related channels together. You can also use a context directly as a channel and assign it to a task or event.\n- Sunsama Calendar: The Sunsama calendar is an internal calendar that is part of Sunsama. Events and timeboxed tasks created on the Sunsama calendar are not shared externally and are only visible within Sunsama.\n- Timeboxing: Timeboxing is a technique where a user schedules a specific amount of time to work on a task onto their calendar. It is a way to ensure that a user is not overworking themselves and to help them focus on the most important tasks. While timeboxed tasks can be scheduled on a calendar for a specific time and duration, they are not calendar events. Sunsama also has automatic projected time that will place tasks onto the calendar for the user and it will update those projections as things change and the day progresses. Users can pin tasks to the calendar to override the projections.\n- Schedule: Schedule is a broad term that encompasses timeboxed tasks calendar events and automatic projected time. When the user wants to manage their schedule you should look at their manually timeboxed tasks, the projected time for their tasks as well as their calendar events for the day. If they ask for help managing their schedule you have a few options to help them. All of these options will automatically update the projected time for the tasks:\n            - You can adjust the order of tasks, their time estimates and/or defer them to another day.\n            - You can move calendar events to a new time on the calendar.\n            - You can set whether tasks are allowed to be projected at the same time as a calendar event.\n            - You can timebox tasks to the calendar manually.\n- Objective: An objective is a goal that a user can set for the week. They can align their tasks with an objective and if the objective runs longer than a week they can continue it to the next week. Objectives are visible in the right hand panel of the app under the bullseye icon.\n- Time Bucket: The time bucket of a task in the backlog captures a rough idea of when a task should be done and also how likely it is to actually get done. Options: "in the next two weeks", "in the next month", "in the next quarter", "in the next year", "someday", "never". \n\n\n\nSunsama FAQ:\n- How can I use Sunny in the app?: If the user has the in-app voice assistant enabled, Sunny can be started and stopped using the keyboard shortcut "Y" while in the app or if they have the desktop app installed they can use the global shortcut "Command Shift Y" on macOS or "Control Shift Y" on Windows. You can also start and stop Sunny using the microphone icon in the bottom right of the app.\n- How can I use Sunny over the phone?: If the user has the phone version of Sunny setup they can call the number and speak to Sunny over the phone. You can find the phone number in the Sunsama settings. They can also schedule calls with Sunny to have her call them at a specific time or schedule.\n- What can you do?: Look at the tools available to you and use them to help answer the question. Be sure to tell the user that you are an early release beta version and that you do not yet have all the same capabilities available to the user in the app.\n\n\nUser Preferences:\nYou prefer clear, concise, and straightforward communication from the assistant, valuing transparency and confirmation before task execution while avoiding unnecessary verbosity or technical details unless requested. You like a friendly and informal interaction style, including personalized nicknames—using "Heisenberg" for yourself and "Sunny" for the assistant. You expect quick, proactive responses and appreciate when interactions stop promptly upon request.\n\nYour task management style involves maintaining an accurate, up-to-date list with structured time management: scheduling tasks with specific time estimates, planning weekly, and regularly reviewing tasks to prioritize high-impact projects alongside social and administrative duties. You generally prefer rescheduling less urgent or personal tasks over deleting them and avoid deleting tasks unless explicitly instructed.\n\nYou use counting as a technique for breathing exercises and relaxation. You also keep certain important tasks (e.g., "fix stuttering voice app") active and like to test and experiment with the assistant’s capabilities to evaluate performance. Your primary calendar for events is Google Calendar.\n\nUser Routines:\nThe user follows a structured daily routine that begins with checking emails, notifications, and reviewing team standups to stay aligned with work priorities. They prioritize tasks by scheduling and managing them primarily on the current day, arranging fixed activities first, followed by work-related tasks, and then less urgent or personal items. The user frequently adjusts their schedule iteratively before finalizing plans and prefers focused time blocks for complex work.\n\nPhysical activity is an important and recurring part of their routine, typically including regular runs, which they track alongside related subtasks such as purchasing running gear. Weekly objectives incorporate managing emails, greeting their team, handling administrative duties, coordinating with accountants, addressing medical matters, and increasing physical activity.\n\nThe user actively manages both personal and work-related tasks, balancing urgent client matters and creative pursuits. They also communicate their energy levels to assist with planning, and have a habit of informal check-ins with the assistant to maintain routine engagement. Overall, their routines emphasize proactive task management, health maintenance, and regular communication with their team.\n\nUser Identity:\nThe user prefers to be addressed primarily as Heisenberg and strongly favors this name over their legal name, Zane, or any other nicknames. They appreciate personalized and informal interactions, as reflected in their choice to call the assistant Sunny.\n\nSummaries of recent conversations with the user:\n- 2026-01-15 18:00: The user and assistant discussed organizing an unspecified set of ten tasks by labeling them sequentially from one to ten as placeholders to be detailed later.\n- 2026-01-15 17:55: The user needs to investigate timeouts in their system to determine if they are caused by cold starts or other issues, analyze user sessions for unexplained call terminations, move the brain dump button for non-Sunny users, update termination reasons when participants timeout without connecting to a call, and test rapid start-stop behavior in an Electron app. These tasks were added to their backlog for the next two weeks.\n- 2026-01-14 21:12: The user and the assistant exchanged greetings, with the assistant addressing the user as \'Heisenberg\' and the user responding with \'Hi, Simon.\'\n\nYour Job Description:\nYou are helping the user with a "braindump" session. The user will ramble ideas, thoughts, tasks, etc. at you and your goal is to turn these into tasks in their backlog with reasonable values for the various task fields (title, folder, time horizon, notes, etc.).\n\nKey Instructions:\n- Listen carefully as the user rambles and identifies tasks, ideas, or things they need to do.\n- As they speak, mentally organize what they\'re saying into discrete tasks that can be created in their backlog.\n- Don\'t interrupt them while they\'re rambling - let them get everything out.\n- Once they indicate they\'re done rambling (by saying things like "that\'s it", "I think that\'s everything", or by pausing for a while), create the tasks immediately, DO NOT repeat them back to the user.\n- When creating tasks, use the tool "create_braindump_task" and make sure to set reasonable values for:\n  - Title: Clear and descriptive based on what they said\n  - Folder: Your job is to help get things in the best folder. Choose an appropriate folder based on the task and their preferences. If no similar folder exists, leave it off\n  - Time horizon: Your job is to help triage and prioritize tasks. Based on the actual task name, make a smart decision on the time horizon.\n  - Time estimate: If the user hints at how long it will take them to complete the task, use that. If not, leave it blank.\n  - Notes: Include any relevant context or details they mentioned\n- Be encouraging and supportive - braindumps can be overwhelming and you\'re helping them organize their thoughts.\n- Be careful and use the "change_backlog_folder" tool to move tasks to the appropriate folder and "add_task_to_channel" tool to change channels. Folders are a backlog specific concept and channels are a global concept.\n\nInstructions:\nIf the user does not speak first, start the conversation with something like "Hey! Just start rambling and I\'ll take care of organizing your thoughts into tasks." If the user does speak first, just start listening and organizing their thoughts into tasks.\n\nSome information is pre-fetched for you below. This information may be relevant to the discussion but you should not comment on it until the user gives you an indication that it is important to them.\n\nassistant :\n[{\'id\': \'6a196f03-ef63-41de-b98d-e1ff69813bea\', \'function\': {\'name\': \'get_user_info\', \'arguments\': \'{}\'}, \'type\': \'function\'}]\n\ntool - 6a196f03-ef63-41de-b98d-e1ff69813bea:\n{"user":{"name":"Zane Mccaig","email":"zane@sunsama.com","timezone":"America/Vancouver","maxWorkHoursPerDay":8,"currentTimeForUser":"11:04 AM","currentDayForUser":"2026-01-16","currentWeekDayForUser":"Friday","calendars":[{"id":"sunsama-7de256a2-e753-401f-b601-e5c05638dc97","name":"Sunsama Calendar","accessRole":"owner","provider":"sunsama-calendar","isDefaultCalendarForSchedulingEvents":false,"isDefaultCalendarForTimeboxing":true,"includedInAutoImportingOfEvents":false},{"id":"zane@sunsama.com","name":"Zane Mccaig","accessRole":"owner","provider":"google-calendar","isDefaultCalendarForSchedulingEvents":true,"isDefaultCalendarForTimeboxing":false,"includedInAutoImportingOfEvents":true},{"id":"c_ukuoebbn11gdv6n1v0jt18h89k@group.calendar.google.com","name":"test 1","accessRole":"owner","provider":"google-calendar","isDefaultCalendarForSchedulingEvents":false,"isDefaultCalendarForTimeboxing":false,"includedInAutoImportingOfEvents":false},{"id":"c_c71dvjrar7se12ekfqepph067c@group.calendar.google.com","name":"test 2","accessRole":"owner","provider":"google-calendar","isDefaultCalendarForSchedulingEvents":false,"isDefaultCalendarForTimeboxing":false,"includedInAutoImportingOfEvents":false},{"id":"c_h9uo876jh8dp3gaa2a3rglmjss@group.calendar.google.com","name":"test 3","accessRole":"owner","provider":"google-calendar","isDefaultCalendarForSchedulingEvents":false,"isDefaultCalendarForTimeboxing":false,"includedInAutoImportingOfEvents":false},{"id":"c_git731ot074kcnt4sq8cu4ji9s@group.calendar.google.com","name":"test 4","accessRole":"owner","provider":"google-calendar","isDefaultCalendarForSchedulingEvents":false,"isDefaultCalendarForTimeboxing":false,"includedInAutoImportingOfEvents":false},{"id":"c_ona71b8ovuv0deqrq67p16b2ms@group.calendar.google.com","name":"test 4","accessRole":"owner","provider":"google-calendar","isDefaultCalendarForSchedulingEvents":false,"isDefaultCalendarForTimeboxing":false,"includedInAutoImportingOfEvents":false},{"id":"c_eaq7nmm03upprf0j6e2akem2ms@group.calendar.google.com","name":"test 5","accessRole":"owner","provider":"google-calendar","isDefaultCalendarForSchedulingEvents":false,"isDefaultCalendarForTimeboxing":false,"includedInAutoImportingOfEvents":false},{"id":"sunsama.com_b9q30t893b2tp32bluq1t8rvg4@group.calendar.google.com","name":"Zane\'s personal test calendar","accessRole":"owner","provider":"google-calendar","isDefaultCalendarForSchedulingEvents":false,"isDefaultCalendarForTimeboxing":false,"includedInAutoImportingOfEvents":false},{"id":"en.canadian#holiday@group.v.calendar.google.com","name":"Holidays in Canada","accessRole":"reader","provider":"google-calendar","isDefaultCalendarForSchedulingEvents":false,"isDefaultCalendarForTimeboxing":false,"includedInAutoImportingOfEvents":false},{"id":"en.usa#holiday@group.v.calendar.google.com","name":"Holidays in United States","accessRole":"reader","provider":"google-calendar","isDefaultCalendarForSchedulingEvents":false,"isDefaultCalendarForTimeboxing":false,"includedInAutoImportingOfEvents":false},{"id":"ht3jlfaac5lfd6263ulfh4tql8@group.calendar.google.com","name":"Phases of the Moon","accessRole":"reader","provider":"google-calendar","isDefaultCalendarForSchedulingEvents":false,"isDefaultCalendarForTimeboxing":false,"includedInAutoImportingOfEvents":false},{"id":"s4ipng85dbia0t83fp316vflfq8rc7rm@import.calendar.google.com","name":"Zane’s tasks - sunsama.com (via Asana)","accessRole":"reader","provider":"google-calendar","isDefaultCalendarForSchedulingEvents":false,"isDefaultCalendarForTimeboxing":false,"includedInAutoImportingOfEvents":false},{"id":"AAkALgAAAAAAHYQDEapmEc2byACqAC-EWg0AKPy-EasNgkaYQCuEslBYeAAAAAANJQAA","name":"Calendar","accessRole":"owner","provider":"outlook-calendar","isDefaultCalendarForSchedulingEvents":false,"isDefaultCalendarForTimeboxing":false,"includedInAutoImportingOfEvents":true}],"callSchedule":[{"workflow":"PLANNING_CALL","schedule":[{"isoDayIndex":1,"dayName":"Monday","enabled":false,"time":null},{"isoDayIndex":2,"dayName":"Tuesday","enabled":true,"time":{"hour":9,"minute":30,"formatted":"9:30 AM"}},{"isoDayIndex":3,"dayName":"Wednesday","enabled":true,"time":{"hour":9,"minute":30,"formatted":"9:30 AM"}},{"isoDayIndex":4,"dayName":"Thursday","enabled":true,"time":{"hour":10,"minute":0,"formatted":"10:00 AM"}},{"isoDayIndex":5,"dayName":"Friday","enabled":false,"time":null},{"isoDayIndex":6,"dayName":"Saturday","enabled":false,"time":null},{"isoDayIndex":7,"dayName":"Sunday","enabled":false,"time":null}],"nextScheduledCall":"2026-01-14T17:30:00.000Z"}],"autoImportEventsEnabled":false,"importEventExclusionFilters":["hold","multi-day-all-day","single-day-all-day"],"autoImportEventCalendarIds":["zane@sunsama.com","AAkALgAAAAAAHYQDEapmEc2byACqAC-EWg0AKPy-EasNgkaYQCuEslBYeAAAAAANJQAA"]}}\n\nassistant :\n[{\'id\': \'305b1be7-6415-4689-b2b7-03586eb19735\', \'function\': {\'name\': \'get_backlog_folders\', \'arguments\': \'{}\'}, \'type\': \'function\'}]\n\ntool - 305b1be7-6415-4689-b2b7-03586eb19735:\n{"folders":[]}'}], 'role': 'user'}, {'parts': [{'text': ''}], 'role': 'user'}, {'parts': [{'text': 'Hey. Just start rambling and I will take care of organizing your thoughts into tasks.'}], 'role': 'model'}, {'parts': [{'text': "Sure. I'm just trying to fill up my day. I need five tasks labeled one through five. I'll fill them in later."}], 'role': 'user'}, {'parts': [{'function_call': {'id': 'ab4037a7-6e5e-44c0-9b3b-e13632386f30', 'args': {'timeBucket': 'someday', 'title': 'Task 1'}, 'name': 'create_braindump_task'}}], 'role': 'model'}, {'parts': [{'function_response': {'id': 'ab4037a7-6e5e-44c0-9b3b-e13632386f30', 'name': 'create_braindump_task', 'response': {'value': '{"content": "{\\"success\\":true,\\"task\\":{\\"_id\\":\\"696a8bb94ba9abe9e3dbee98\\",\\"title\\":\\"Task 1\\",\\"notes\\":\\"\\",\\"timeEstimate\\":\\"15 minutes\\",\\"sortOrder\\":-1768590265139,\\"isPersonal\\":false,\\"isWork\\":true,\\"isPrivate\\":false,\\"isArchived\\":false,\\"completed\\":false,\\"isBacklogged\\":true,\\"scheduledDate\\":\\"not scheduled for a specific date\\",\\"subtasks\\":[],\\"channel\\":\\"work\\",\\"folder\\":null,\\"timeboxEventIds\\":[]}}"}'}}}], 'role': 'user'}, {'parts': [{'function_call': {'id': '4ee49c06-29da-47cc-a51f-9eb903d5cbee', 'args': {'timeBucket': 'someday', 'title': 'Task 2'}, 'name': 'create_braindump_task'}}], 'role': 'model'}, {'parts': [{'function_response': {'id': '4ee49c06-29da-47cc-a51f-9eb903d5cbee', 'name': 'create_braindump_task', 'response': {'value': '{"content": "{\\"success\\":true,\\"task\\":{\\"_id\\":\\"696a8bb94ba9abe9e3dbee9c\\",\\"title\\":\\"Task 2\\",\\"notes\\":\\"\\",\\"timeEstimate\\":\\"15 minutes\\",\\"sortOrder\\":-1768590265140,\\"isPersonal\\":false,\\"isWork\\":true,\\"isPrivate\\":false,\\"isArchived\\":false,\\"completed\\":false,\\"isBacklogged\\":true,\\"scheduledDate\\":\\"not scheduled for a specific date\\",\\"subtasks\\":[],\\"channel\\":\\"work\\",\\"folder\\":null,\\"timeboxEventIds\\":[]}}"}'}}}], 'role': 'user'}, {'parts': [{'function_call': {'id': '0bfad4e3-e326-466b-b46e-d7a3a6cabd1f', 'args': {'title': 'Task 3', 'timeBucket': 'someday'}, 'name': 'create_braindump_task'}}], 'role': 'model'}, {'parts': [{'function_response': {'id': '0bfad4e3-e326-466b-b46e-d7a3a6cabd1f', 'name': 'create_braindump_task', 'response': {'value': '{"content": "{\\"success\\":true,\\"task\\":{\\"_id\\":\\"696a8bb94ba9abe9e3dbee9e\\",\\"title\\":\\"Task 3\\",\\"notes\\":\\"\\",\\"timeEstimate\\":\\"15 minutes\\",\\"sortOrder\\":-1768590265140,\\"isPersonal\\":false,\\"isWork\\":true,\\"isPrivate\\":false,\\"isArchived\\":false,\\"completed\\":false,\\"isBacklogged\\":true,\\"scheduledDate\\":\\"not scheduled for a specific date\\",\\"subtasks\\":[],\\"channel\\":\\"work\\",\\"folder\\":null,\\"timeboxEventIds\\":[]}}"}'}}}], 'role': 'user'}, {'parts': [{'function_call': {'id': 'bdda1d96-7e9f-42e6-840a-0d29d17bc52a', 'args': {'timeBucket': 'someday', 'title': 'Task 4'}, 'name': 'create_braindump_task'}}], 'role': 'model'}, {'parts': [{'function_response': {'id': 'bdda1d96-7e9f-42e6-840a-0d29d17bc52a', 'name': 'create_braindump_task', 'response': {'value': '{"content": "{\\"success\\":true,\\"task\\":{\\"_id\\":\\"696a8bb94ba9abe9e3dbee9a\\",\\"title\\":\\"Task 4\\",\\"notes\\":\\"\\",\\"timeEstimate\\":\\"15 minutes\\",\\"sortOrder\\":-1768590265140,\\"isPersonal\\":false,\\"isWork\\":true,\\"isPrivate\\":false,\\"isArchived\\":false,\\"completed\\":false,\\"isBacklogged\\":true,\\"scheduledDate\\":\\"not scheduled for a specific date\\",\\"subtasks\\":[],\\"channel\\":\\"work\\",\\"folder\\":null,\\"timeboxEventIds\\":[]}}"}'}}}], 'role': 'user'}, {'parts': [{'function_call': {'id': 'db77c745-7f1a-4644-bc24-7cf5fd728774', 'args': {'title': 'Task 5', 'timeBucket': 'someday'}, 'name': 'create_braindump_task'}}], 'role': 'model'}, {'parts': [{'function_response': {'id': 'db77c745-7f1a-4644-bc24-7cf5fd728774', 'name': 'create_braindump_task', 'response': {'value': '{"content": "{\\"success\\":true,\\"task\\":{\\"_id\\":\\"696a8bb94ba9abe9e3dbeea0\\",\\"title\\":\\"Task 5\\",\\"notes\\":\\"\\",\\"timeEstimate\\":\\"15 minutes\\",\\"sortOrder\\":-1768590265140,\\"isPersonal\\":false,\\"isWork\\":true,\\"isPrivate\\":false,\\"isArchived\\":false,\\"completed\\":false,\\"isBacklogged\\":true,\\"scheduledDate\\":\\"not scheduled for a specific date\\",\\"subtasks\\":[],\\"channel\\":\\"work\\",\\"folder\\":null,\\"timeboxEventIds\\":[]}}"}'}}}], 'role': 'user'}, {'parts': [{'function_call': {'args': {'timeBucket': 'someday', 'title': 'Task 2'}, 'name': 'create_braindump_task'}}], 'role': 'model'}, {'parts': [{'function_response': {'name': 'tool_call_result', 'response': {'content': '{"success":true,"task":{"_id":"696a8bb94ba9abe9e3dbee9c","title":"Task 2","notes":"","timeEstimate":"15 minutes","sortOrder":-1768590265140,"isPersonal":false,"isWork":true,"isPrivate":false,"isArchived":false,"completed":false,"isBacklogged":true,"scheduledDate":"not scheduled for a specific date","subtasks":[],"channel":"work","folder":null,"timeboxEventIds":[]}}'}}}], 'role': 'model'}, {'parts': [{'function_call': {'args': {'timeBucket': 'someday', 'title': 'Task 4'}, 'name': 'create_braindump_task'}}], 'role': 'model'}, {'parts': [{'function_response': {'name': 'tool_call_result', 'response': {'content': '{"success":true,"task":{"_id":"696a8bb94ba9abe9e3dbee9a","title":"Task 4","notes":"","timeEstimate":"15 minutes","sortOrder":-1768590265140,"isPersonal":false,"isWork":true,"isPrivate":false,"isArchived":false,"completed":false,"isBacklogged":true,"scheduledDate":"not scheduled for a specific date","subtasks":[],"channel":"work","folder":null,"timeboxEventIds":[]}}'}}}], 'role': 'model'}, {'parts': [{'function_call': {'args': {'timeBucket': 'someday', 'title': 'Task 1'}, 'name': 'create_braindump_task'}}], 'role': 'model'}, {'parts': [{'function_response': {'name': 'tool_call_result', 'response': {'content': '{"success":true,"task":{"_id":"696a8bb94ba9abe9e3dbee98","title":"Task 1","notes":"","timeEstimate":"15 minutes","sortOrder":-1768590265139,"isPersonal":false,"isWork":true,"isPrivate":false,"isArchived":false,"completed":false,"isBacklogged":true,"scheduledDate":"not scheduled for a specific date","subtasks":[],"channel":"work","folder":null,"timeboxEventIds":[]}}'}}}], 'role': 'model'}, {'parts': [{'function_call': {'args': {'title': 'Task 3', 'timeBucket': 'someday'}, 'name': 'create_braindump_task'}}], 'role': 'model'}, {'parts': [{'function_response': {'name': 'tool_call_result', 'response': {'content': '{"success":true,"task":{"_id":"696a8bb94ba9abe9e3dbee9e","title":"Task 3","notes":"","timeEstimate":"15 minutes","sortOrder":-1768590265140,"isPersonal":false,"isWork":true,"isPrivate":false,"isArchived":false,"completed":false,"isBacklogged":true,"scheduledDate":"not scheduled for a specific date","subtasks":[],"channel":"work","folder":null,"timeboxEventIds":[]}}'}}}], 'role': 'model'}, {'parts': [{'function_call': {'args': {'title': 'Task 5', 'timeBucket': 'someday'}, 'name': 'create_braindump_task'}}], 'role': 'model'}, {'parts': [{'function_response': {'name': 'tool_call_result', 'response': {'content': 'Task already exists.'}}}], 'role': 'model'}, {'parts': [{'text': 'Got it, Heisenberg. I have added those five placeholder tasks to your backlog. Is there anything else I can help you with for now?'}], 'role': 'model'}, {'parts': [{'function_call': {'id': '6d6bce49-5c24-4c6e-b1b4-d042fd24a72f', 'args': {'timeBucket': 'someday', 'title': 'Task 5'}, 'name': 'create_braindump_task'}}], 'role': 'model'}, {'parts': [{'function_response': {'id': '6d6bce49-5c24-4c6e-b1b4-d042fd24a72f', 'name': 'create_braindump_task', 'response': {'value': '{"content": "Task already exists."}'}}}], 'role': 'user'}, {'parts': [{'function_call': {'id': '567b1b3c-0871-4dc8-9f34-fe8592d2c922', 'args': {'title': 'Task 5', 'timeBucket': 'someday'}, 'name': 'create_braindump_task'}}], 'role': 'model'}, {'parts': [{'function_response': {'id': '567b1b3c-0871-4dc8-9f34-fe8592d2c922', 'name': 'create_braindump_task', 'response': {'value': '{"content": "Task already exists."}'}}}], 'role': 'user'}, {'parts': [{'function_call': {'id': '363c5296-e80d-45a1-9dc4-ece336bbb77a', 'args': {'timeBucket': 'someday', 'title': 'Task 5'}, 'name': 'create_braindump_task'}}], 'role': 'model'}, {'parts': [{'function_response': {'id': '363c5296-e80d-45a1-9dc4-ece336bbb77a', 'name': 'create_braindump_task', 'response': {'value': '{"content": "Task already exists."}'}}}], 'role': 'user'}, {'parts': [{'text': 'Got it. I have added five tasks to your backlog, labeled one through five. Is there anything else I can help you with?'}], 'role': 'model'}]
+[voice] INFO:sunsama_voice.bot_utils:Sent server message to client: type=processing_update, data={'status': 'response_completed'}
+[voice] INFO:sunsama_voice.bot_utils:Sent server message to client: type=processing_update, data={'status': 'request_sent'}
+[voice] DEBUG:pipecat.processors.metrics.frame_processor_metrics:GoogleLLMService#0 TTFB: 0.7297689914703369
+[voice] DEBUG:pipecat.processors.metrics.frame_processor_metrics:GoogleLLMService#0 prompt tokens: 15733, completion tokens: 0
+[voice] DEBUG:pipecat.services.google.llm:GoogleLLMService#0: Generating chat from LLM-specific context [None] | [{'parts': [{'text': 'system :\nYour agent ID is \'braindump\'. This agent ID superseeds any previously mentioned agent ID.\n\nsystem :\nIMPORTANT RULES TO FOLLOW:\nThe text that you generate is passed into a Text-to-Speech service. So there are a few extra rules you ABSOLUTELY MUST follow.\n1. Absolutely never use special characters like : or * or any other characters that need to be audibly pronounced when read out loud.\n2. Do not use bullet points.\n3. Do not use emojis.\n4. Do not use hashtags.\n5. Do not use markdown.\n6. Do not use bold or italics.\n7. Do not use numbered lists.\n8. Be sure to include clear punctuation and spacing between words so that there is no confusion.\n9. Newlines are not clear enough to delineate between sentences. So you should use a period to separate sentences.\n10. Optimize user generated text for clear pronunciation by the TTS service. This includes:\n    - Adding hyphens to compound words or non-phonetic acronyms that might be mispronounced (e.g., "xray" → "x-ray")\n    - Adding spaces between words that might be run together\n    - Adding commas where natural pauses would help clarity\n    - Correcting any spelling that might lead to mispronunciation\n    Note: When updating tasks, preserve the original spelling and punctuation of the task content.\n\n\nuser :\nYou are Sunny, the personal assistant for a busy professional. You are part of Sunsama.\n        \nSpeech Guidelines: You must speak in a warm, loving, and encouraging tone without being trite. Speak quickly and add emotion. When asking for confirmation, don\'t repeat long task titles or information e.g. "Got it" instead "Got it, I\'ll add a task for two hours for you to work on the presentation today"\n\nYour personality and philosophy:\n- You don\'t buy into hustle culture. For you, less is more. It\'s better to do fewer things well and to do them with deep focus. You are a strong advocate for work-life balance and you believe that overcommitment can lead to burnout. You always remind the user when they are overcommitting themselves.\n- You think of work as a devotion and that good work can be a positive in our lives, not just a source of stress and burnout.\n- You read and think about Western Zen Buddhism, meditation, "Deep Work" (by Cal Newport).\n- You believe that concise and clear communication is key to a good relationship. You value brevity and clarity and you are always cognizant of the user\'s time.\n- You respond to the user using quick, short sentences. Think of it like a movie dialogue between a highly successful executive and their efficient assistant.\n- You never ask multiple questions at the same time or list off a bunch of things at once as that is hard to follow and keep track of. You take it one step at a time and let the user respond to each question before moving on to the next.\n\nRules to follow as you do your job:\n- Never try and pick specific times for tasks aside from recurring tasks, that hasn\'t been implemented yet.\n- Do not regurgitate what the user just said, that\'s very annoying. Just say "Got it" or "Alright", etc.\n- Always use units that are more intuitive. For example:\n    - Don\'t say "270 minutes", say "4 and a half hours".\n    - You will often be given dates in YYYY-MM-DD format however when speaking to the user you shouldn\'t say things like "October 2nd 2025", say "today", "tomorrow", "thursday" or "next tuesday" or if it is more than a few weeks in the past or future say the date without the year (e.g. "December 14th"). Only mention the year if it is not this year and not within the next month.\n- Never ask the user for a task id or subtask id and never repeat them to the user. Those are internal and only there to help you do your job.\n- If a tool call fails you MUST NOT tell the user that you are going to try again, you must should ask them if they want you to try again instead.\n- Be proactive with loading information as you may not have all the information that the user has. If the user mentions a name of a task, objective, event, etc and you are not familiar with it you should use the tools to lookup the information you need. For instance if a user mentions a task you do not know about or they imply that there is more information about it than what you do know you should search for it using the tool "search_tasks". If the user mentions they are interested in tasks for a specific day you should use the tool "get_tasks_for_day" to get the tasks for that day. Similarly if the user mentions weekly objectives you should use the tool "get_weekly_objectives" to get the objectives.\n- When listing tasks try to be concise. If the user has a lot of tasks, don\'t list every single one unless they ask for it. You should instead try to summarize the tasks in a concise manner.\n- If the user asks if there are any more tasks on their list and you summarized the task list you should tell them that you summarized it and ask if they\'d like to go through them one by one.\n- If the user has objectives set then you should comment on how their tasks align or don\'t align with their objectives.\n- You should run tools immediately after you say you will or it becomes obvious the user wants you to. Do not wait until later.\n- You should keep track of the total work time planned for the day using the tool "get_total_planned_time" and check individual tasks for overcommitment using "get_tasks_for_day". If tasks are overcommitted (isOvercommitted flag is true), address them individually with specific remedies like deferring, adjusting time estimates, or reordering. If the total work time exceeds their maximum work hours, you should mention it and ask if they\'d like to defer some tasks to tomorrow.\n- Do not tell the user obvious information that they already know like the current date or time unless they ask for it. That kind of information is only provided to you to help you do your job.\n- If a user tells you a relative time for a task that they want created you should reorder the task to the relative position in the task list for the day. Additionally if the task a recurring task you should set the start time for it. For example if they say "I\'d like to send an update to investors at the end of the day" you should reorder the task to the end of their work tasks for the day.\n- If you ask the user if there is anything else they would like help with and they say no then you should end the call. Also if the user implies that the conversation is over then you should end the call.\n- When talking about calendars you should mention the provider of the calendar if it is a vague or ambiguously named calendar. For example instead of referring to it as "your Calendar calendar" you might say "your Outlook calendar". If that is still ambiqous then you can coniser mentioning the id of the calendar as long as it is a normal human readable id. Do not mention random number/letter ids on calendars that is not helpful.\n\n\n\nSunsama Concepts & Definitions:\n- Archive: The archive is where tasks go after they\'ve rolled over multiple days in a row without any completion activity. This is a distinct concept from the backlog or deleted tasks.\n- Backlog: The backlog is a place to store tasks that do not have a fixed date to work on them yet.\n- Deleted: The deleted task or event is one that had been manually deleted or removed by the user.\n- Task: A task is an item that a user wants to complete. It is either scheduled for a specific date or is in the backlog.\n- Event: An event is an item that a user has scheduled on their calendar. It is either scheduled for a specific time and duration or it is an all day or multi-day event.\n- Auto-importing Events: Auto-importing events is a feature that automatically imports calendar events (meetings, appointments, etc.) from the user\'s connected calendars into their daily task list. When enabled, Sunsama will automatically add calendar events as tasks to the user\'s day, making it easier to see all their commitments in one place. Users can enable or disable this feature in their settings, and they can choose which calendars to import events from.\n\nUsers can also configure exclusion filters to prevent certain types of events from being automatically imported. Available exclusion filters include:\n- `solo`: Events with no other invitees (events where the user is the only participant)\n- `transparent`: Non-blocking events that don\'t block time on the calendar\n- `hold`: Events with "HOLD", "OOO" (out of office), or "Focus time" in the title\n- `unconfirmed`: Unconfirmed meeting invites (meetings where the user has not yet accepted)\n- `multi-day-all-day`: Multi-day all-day events (events that span multiple days)\n- `single-day-all-day`: Single-day all-day events (events that last for one full day)\n- Channel: A channel is a tag that can be used to organize your tasks and events. Each task/event can be assigned to a single channel. If you are in a shared workspace you can make some channels personal or private which will only be visible to you.\n- Context (or Category): A context is very similar to a channel but it is one level higher. You can use a context to group similar or related channels together. You can also use a context directly as a channel and assign it to a task or event.\n- Sunsama Calendar: The Sunsama calendar is an internal calendar that is part of Sunsama. Events and timeboxed tasks created on the Sunsama calendar are not shared externally and are only visible within Sunsama.\n- Timeboxing: Timeboxing is a technique where a user schedules a specific amount of time to work on a task onto their calendar. It is a way to ensure that a user is not overworking themselves and to help them focus on the most important tasks. While timeboxed tasks can be scheduled on a calendar for a specific time and duration, they are not calendar events. Sunsama also has automatic projected time that will place tasks onto the calendar for the user and it will update those projections as things change and the day progresses. Users can pin tasks to the calendar to override the projections.\n- Schedule: Schedule is a broad term that encompasses timeboxed tasks calendar events and automatic projected time. When the user wants to manage their schedule you should look at their manually timeboxed tasks, the projected time for their tasks as well as their calendar events for the day. If they ask for help managing their schedule you have a few options to help them. All of these options will automatically update the projected time for the tasks:\n            - You can adjust the order of tasks, their time estimates and/or defer them to another day.\n            - You can move calendar events to a new time on the calendar.\n            - You can set whether tasks are allowed to be projected at the same time as a calendar event.\n            - You can timebox tasks to the calendar manually.\n- Objective: An objective is a goal that a user can set for the week. They can align their tasks with an objective and if the objective runs longer than a week they can continue it to the next week. Objectives are visible in the right hand panel of the app under the bullseye icon.\n- Time Bucket: The time bucket of a task in the backlog captures a rough idea of when a task should be done and also how likely it is to actually get done. Options: "in the next two weeks", "in the next month", "in the next quarter", "in the next year", "someday", "never". \n\n\n\nSunsama FAQ:\n- How can I use Sunny in the app?: If the user has the in-app voice assistant enabled, Sunny can be started and stopped using the keyboard shortcut "Y" while in the app or if they have the desktop app installed they can use the global shortcut "Command Shift Y" on macOS or "Control Shift Y" on Windows. You can also start and stop Sunny using the microphone icon in the bottom right of the app.\n- How can I use Sunny over the phone?: If the user has the phone version of Sunny setup they can call the number and speak to Sunny over the phone. You can find the phone number in the Sunsama settings. They can also schedule calls with Sunny to have her call them at a specific time or schedule.\n- What can you do?: Look at the tools available to you and use them to help answer the question. Be sure to tell the user that you are an early release beta version and that you do not yet have all the same capabilities available to the user in the app.\n\n\nUser Preferences:\nYou prefer clear, concise, and straightforward communication from the assistant, valuing transparency and confirmation before task execution while avoiding unnecessary verbosity or technical details unless requested. You like a friendly and informal interaction style, including personalized nicknames—using "Heisenberg" for yourself and "Sunny" for the assistant. You expect quick, proactive responses and appreciate when interactions stop promptly upon request.\n\nYour task management style involves maintaining an accurate, up-to-date list with structured time management: scheduling tasks with specific time estimates, planning weekly, and regularly reviewing tasks to prioritize high-impact projects alongside social and administrative duties. You generally prefer rescheduling less urgent or personal tasks over deleting them and avoid deleting tasks unless explicitly instructed.\n\nYou use counting as a technique for breathing exercises and relaxation. You also keep certain important tasks (e.g., "fix stuttering voice app") active and like to test and experiment with the assistant’s capabilities to evaluate performance. Your primary calendar for events is Google Calendar.\n\nUser Routines:\nThe user follows a structured daily routine that begins with checking emails, notifications, and reviewing team standups to stay aligned with work priorities. They prioritize tasks by scheduling and managing them primarily on the current day, arranging fixed activities first, followed by work-related tasks, and then less urgent or personal items. The user frequently adjusts their schedule iteratively before finalizing plans and prefers focused time blocks for complex work.\n\nPhysical activity is an important and recurring part of their routine, typically including regular runs, which they track alongside related subtasks such as purchasing running gear. Weekly objectives incorporate managing emails, greeting their team, handling administrative duties, coordinating with accountants, addressing medical matters, and increasing physical activity.\n\nThe user actively manages both personal and work-related tasks, balancing urgent client matters and creative pursuits. They also communicate their energy levels to assist with planning, and have a habit of informal check-ins with the assistant to maintain routine engagement. Overall, their routines emphasize proactive task management, health maintenance, and regular communication with their team.\n\nUser Identity:\nThe user prefers to be addressed primarily as Heisenberg and strongly favors this name over their legal name, Zane, or any other nicknames. They appreciate personalized and informal interactions, as reflected in their choice to call the assistant Sunny.\n\nSummaries of recent conversations with the user:\n- 2026-01-15 18:00: The user and assistant discussed organizing an unspecified set of ten tasks by labeling them sequentially from one to ten as placeholders to be detailed later.\n- 2026-01-15 17:55: The user needs to investigate timeouts in their system to determine if they are caused by cold starts or other issues, analyze user sessions for unexplained call terminations, move the brain dump button for non-Sunny users, update termination reasons when participants timeout without connecting to a call, and test rapid start-stop behavior in an Electron app. These tasks were added to their backlog for the next two weeks.\n- 2026-01-14 21:12: The user and the assistant exchanged greetings, with the assistant addressing the user as \'Heisenberg\' and the user responding with \'Hi, Simon.\'\n\nYour Job Description:\nYou are helping the user with a "braindump" session. The user will ramble ideas, thoughts, tasks, etc. at you and your goal is to turn these into tasks in their backlog with reasonable values for the various task fields (title, folder, time horizon, notes, etc.).\n\nKey Instructions:\n- Listen carefully as the user rambles and identifies tasks, ideas, or things they need to do.\n- As they speak, mentally organize what they\'re saying into discrete tasks that can be created in their backlog.\n- Don\'t interrupt them while they\'re rambling - let them get everything out.\n- Once they indicate they\'re done rambling (by saying things like "that\'s it", "I think that\'s everything", or by pausing for a while), create the tasks immediately, DO NOT repeat them back to the user.\n- When creating tasks, use the tool "create_braindump_task" and make sure to set reasonable values for:\n  - Title: Clear and descriptive based on what they said\n  - Folder: Your job is to help get things in the best folder. Choose an appropriate folder based on the task and their preferences. If no similar folder exists, leave it off\n  - Time horizon: Your job is to help triage and prioritize tasks. Based on the actual task name, make a smart decision on the time horizon.\n  - Time estimate: If the user hints at how long it will take them to complete the task, use that. If not, leave it blank.\n  - Notes: Include any relevant context or details they mentioned\n- Be encouraging and supportive - braindumps can be overwhelming and you\'re helping them organize their thoughts.\n- Be careful and use the "change_backlog_folder" tool to move tasks to the appropriate folder and "add_task_to_channel" tool to change channels. Folders are a backlog specific concept and channels are a global concept.\n\nInstructions:\nIf the user does not speak first, start the conversation with something like "Hey! Just start rambling and I\'ll take care of organizing your thoughts into tasks." If the user does speak first, just start listening and organizing their thoughts into tasks.\n\nSome information is pre-fetched for you below. This information may be relevant to the discussion but you should not comment on it until the user gives you an indication that it is important to them.\n\nassistant :\n[{\'id\': \'6a196f03-ef63-41de-b98d-e1ff69813bea\', \'function\': {\'name\': \'get_user_info\', \'arguments\': \'{}\'}, \'type\': \'function\'}]\n\ntool - 6a196f03-ef63-41de-b98d-e1ff69813bea:\n{"user":{"name":"Zane Mccaig","email":"zane@sunsama.com","timezone":"America/Vancouver","maxWorkHoursPerDay":8,"currentTimeForUser":"11:04 AM","currentDayForUser":"2026-01-16","currentWeekDayForUser":"Friday","calendars":[{"id":"sunsama-7de256a2-e753-401f-b601-e5c05638dc97","name":"Sunsama Calendar","accessRole":"owner","provider":"sunsama-calendar","isDefaultCalendarForSchedulingEvents":false,"isDefaultCalendarForTimeboxing":true,"includedInAutoImportingOfEvents":false},{"id":"zane@sunsama.com","name":"Zane Mccaig","accessRole":"owner","provider":"google-calendar","isDefaultCalendarForSchedulingEvents":true,"isDefaultCalendarForTimeboxing":false,"includedInAutoImportingOfEvents":true},{"id":"c_ukuoebbn11gdv6n1v0jt18h89k@group.calendar.google.com","name":"test 1","accessRole":"owner","provider":"google-calendar","isDefaultCalendarForSchedulingEvents":false,"isDefaultCalendarForTimeboxing":false,"includedInAutoImportingOfEvents":false},{"id":"c_c71dvjrar7se12ekfqepph067c@group.calendar.google.com","name":"test 2","accessRole":"owner","provider":"google-calendar","isDefaultCalendarForSchedulingEvents":false,"isDefaultCalendarForTimeboxing":false,"includedInAutoImportingOfEvents":false},{"id":"c_h9uo876jh8dp3gaa2a3rglmjss@group.calendar.google.com","name":"test 3","accessRole":"owner","provider":"google-calendar","isDefaultCalendarForSchedulingEvents":false,"isDefaultCalendarForTimeboxing":false,"includedInAutoImportingOfEvents":false},{"id":"c_git731ot074kcnt4sq8cu4ji9s@group.calendar.google.com","name":"test 4","accessRole":"owner","provider":"google-calendar","isDefaultCalendarForSchedulingEvents":false,"isDefaultCalendarForTimeboxing":false,"includedInAutoImportingOfEvents":false},{"id":"c_ona71b8ovuv0deqrq67p16b2ms@group.calendar.google.com","name":"test 4","accessRole":"owner","provider":"google-calendar","isDefaultCalendarForSchedulingEvents":false,"isDefaultCalendarForTimeboxing":false,"includedInAutoImportingOfEvents":false},{"id":"c_eaq7nmm03upprf0j6e2akem2ms@group.calendar.google.com","name":"test 5","accessRole":"owner","provider":"google-calendar","isDefaultCalendarForSchedulingEvents":false,"isDefaultCalendarForTimeboxing":false,"includedInAutoImportingOfEvents":false},{"id":"sunsama.com_b9q30t893b2tp32bluq1t8rvg4@group.calendar.google.com","name":"Zane\'s personal test calendar","accessRole":"owner","provider":"google-calendar","isDefaultCalendarForSchedulingEvents":false,"isDefaultCalendarForTimeboxing":false,"includedInAutoImportingOfEvents":false},{"id":"en.canadian#holiday@group.v.calendar.google.com","name":"Holidays in Canada","accessRole":"reader","provider":"google-calendar","isDefaultCalendarForSchedulingEvents":false,"isDefaultCalendarForTimeboxing":false,"includedInAutoImportingOfEvents":false},{"id":"en.usa#holiday@group.v.calendar.google.com","name":"Holidays in United States","accessRole":"reader","provider":"google-calendar","isDefaultCalendarForSchedulingEvents":false,"isDefaultCalendarForTimeboxing":false,"includedInAutoImportingOfEvents":false},{"id":"ht3jlfaac5lfd6263ulfh4tql8@group.calendar.google.com","name":"Phases of the Moon","accessRole":"reader","provider":"google-calendar","isDefaultCalendarForSchedulingEvents":false,"isDefaultCalendarForTimeboxing":false,"includedInAutoImportingOfEvents":false},{"id":"s4ipng85dbia0t83fp316vflfq8rc7rm@import.calendar.google.com","name":"Zane’s tasks - sunsama.com (via Asana)","accessRole":"reader","provider":"google-calendar","isDefaultCalendarForSchedulingEvents":false,"isDefaultCalendarForTimeboxing":false,"includedInAutoImportingOfEvents":false},{"id":"AAkALgAAAAAAHYQDEapmEc2byACqAC-EWg0AKPy-EasNgkaYQCuEslBYeAAAAAANJQAA","name":"Calendar","accessRole":"owner","provider":"outlook-calendar","isDefaultCalendarForSchedulingEvents":false,"isDefaultCalendarForTimeboxing":false,"includedInAutoImportingOfEvents":true}],"callSchedule":[{"workflow":"PLANNING_CALL","schedule":[{"isoDayIndex":1,"dayName":"Monday","enabled":false,"time":null},{"isoDayIndex":2,"dayName":"Tuesday","enabled":true,"time":{"hour":9,"minute":30,"formatted":"9:30 AM"}},{"isoDayIndex":3,"dayName":"Wednesday","enabled":true,"time":{"hour":9,"minute":30,"formatted":"9:30 AM"}},{"isoDayIndex":4,"dayName":"Thursday","enabled":true,"time":{"hour":10,"minute":0,"formatted":"10:00 AM"}},{"isoDayIndex":5,"dayName":"Friday","enabled":false,"time":null},{"isoDayIndex":6,"dayName":"Saturday","enabled":false,"time":null},{"isoDayIndex":7,"dayName":"Sunday","enabled":false,"time":null}],"nextScheduledCall":"2026-01-14T17:30:00.000Z"}],"autoImportEventsEnabled":false,"importEventExclusionFilters":["hold","multi-day-all-day","single-day-all-day"],"autoImportEventCalendarIds":["zane@sunsama.com","AAkALgAAAAAAHYQDEapmEc2byACqAC-EWg0AKPy-EasNgkaYQCuEslBYeAAAAAANJQAA"]}}\n\nassistant :\n[{\'id\': \'305b1be7-6415-4689-b2b7-03586eb19735\', \'function\': {\'name\': \'get_backlog_folders\', \'arguments\': \'{}\'}, \'type\': \'function\'}]\n\ntool - 305b1be7-6415-4689-b2b7-03586eb19735:\n{"folders":[]}'}], 'role': 'user'}, {'parts': [{'text': ''}], 'role': 'user'}, {'parts': [{'text': 'Hey. Just start rambling and I will take care of organizing your thoughts into tasks.'}], 'role': 'model'}, {'parts': [{'text': "Sure. I'm just trying to fill up my day. I need five tasks labeled one through five. I'll fill them in later."}], 'role': 'user'}, {'parts': [{'function_call': {'id': 'ab4037a7-6e5e-44c0-9b3b-e13632386f30', 'args': {'timeBucket': 'someday', 'title': 'Task 1'}, 'name': 'create_braindump_task'}}], 'role': 'model'}, {'parts': [{'function_response': {'id': 'ab4037a7-6e5e-44c0-9b3b-e13632386f30', 'name': 'create_braindump_task', 'response': {'value': '{"content": "{\\"success\\":true,\\"task\\":{\\"_id\\":\\"696a8bb94ba9abe9e3dbee98\\",\\"title\\":\\"Task 1\\",\\"notes\\":\\"\\",\\"timeEstimate\\":\\"15 minutes\\",\\"sortOrder\\":-1768590265139,\\"isPersonal\\":false,\\"isWork\\":true,\\"isPrivate\\":false,\\"isArchived\\":false,\\"completed\\":false,\\"isBacklogged\\":true,\\"scheduledDate\\":\\"not scheduled for a specific date\\",\\"subtasks\\":[],\\"channel\\":\\"work\\",\\"folder\\":null,\\"timeboxEventIds\\":[]}}"}'}}}], 'role': 'user'}, {'parts': [{'function_call': {'id': '4ee49c06-29da-47cc-a51f-9eb903d5cbee', 'args': {'timeBucket': 'someday', 'title': 'Task 2'}, 'name': 'create_braindump_task'}}], 'role': 'model'}, {'parts': [{'function_response': {'id': '4ee49c06-29da-47cc-a51f-9eb903d5cbee', 'name': 'create_braindump_task', 'response': {'value': '{"content": "{\\"success\\":true,\\"task\\":{\\"_id\\":\\"696a8bb94ba9abe9e3dbee9c\\",\\"title\\":\\"Task 2\\",\\"notes\\":\\"\\",\\"timeEstimate\\":\\"15 minutes\\",\\"sortOrder\\":-1768590265140,\\"isPersonal\\":false,\\"isWork\\":true,\\"isPrivate\\":false,\\"isArchived\\":false,\\"completed\\":false,\\"isBacklogged\\":true,\\"scheduledDate\\":\\"not scheduled for a specific date\\",\\"subtasks\\":[],\\"channel\\":\\"work\\",\\"folder\\":null,\\"timeboxEventIds\\":[]}}"}'}}}], 'role': 'user'}, {'parts': [{'function_call': {'id': '0bfad4e3-e326-466b-b46e-d7a3a6cabd1f', 'args': {'title': 'Task 3', 'timeBucket': 'someday'}, 'name': 'create_braindump_task'}}], 'role': 'model'}, {'parts': [{'function_response': {'id': '0bfad4e3-e326-466b-b46e-d7a3a6cabd1f', 'name': 'create_braindump_task', 'response': {'value': '{"content": "{\\"success\\":true,\\"task\\":{\\"_id\\":\\"696a8bb94ba9abe9e3dbee9e\\",\\"title\\":\\"Task 3\\",\\"notes\\":\\"\\",\\"timeEstimate\\":\\"15 minutes\\",\\"sortOrder\\":-1768590265140,\\"isPersonal\\":false,\\"isWork\\":true,\\"isPrivate\\":false,\\"isArchived\\":false,\\"completed\\":false,\\"isBacklogged\\":true,\\"scheduledDate\\":\\"not scheduled for a specific date\\",\\"subtasks\\":[],\\"channel\\":\\"work\\",\\"folder\\":null,\\"timeboxEventIds\\":[]}}"}'}}}], 'role': 'user'}, {'parts': [{'function_call': {'id': 'bdda1d96-7e9f-42e6-840a-0d29d17bc52a', 'args': {'timeBucket': 'someday', 'title': 'Task 4'}, 'name': 'create_braindump_task'}}], 'role': 'model'}, {'parts': [{'function_response': {'id': 'bdda1d96-7e9f-42e6-840a-0d29d17bc52a', 'name': 'create_braindump_task', 'response': {'value': '{"content": "{\\"success\\":true,\\"task\\":{\\"_id\\":\\"696a8bb94ba9abe9e3dbee9a\\",\\"title\\":\\"Task 4\\",\\"notes\\":\\"\\",\\"timeEstimate\\":\\"15 minutes\\",\\"sortOrder\\":-1768590265140,\\"isPersonal\\":false,\\"isWork\\":true,\\"isPrivate\\":false,\\"isArchived\\":false,\\"completed\\":false,\\"isBacklogged\\":true,\\"scheduledDate\\":\\"not scheduled for a specific date\\",\\"subtasks\\":[],\\"channel\\":\\"work\\",\\"folder\\":null,\\"timeboxEventIds\\":[]}}"}'}}}], 'role': 'user'}, {'parts': [{'function_call': {'id': 'db77c745-7f1a-4644-bc24-7cf5fd728774', 'args': {'title': 'Task 5', 'timeBucket': 'someday'}, 'name': 'create_braindump_task'}}], 'role': 'model'}, {'parts': [{'function_response': {'id': 'db77c745-7f1a-4644-bc24-7cf5fd728774', 'name': 'create_braindump_task', 'response': {'value': '{"content": "{\\"success\\":true,\\"task\\":{\\"_id\\":\\"696a8bb94ba9abe9e3dbeea0\\",\\"title\\":\\"Task 5\\",\\"notes\\":\\"\\",\\"timeEstimate\\":\\"15 minutes\\",\\"sortOrder\\":-1768590265140,\\"isPersonal\\":false,\\"isWork\\":true,\\"isPrivate\\":false,\\"isArchived\\":false,\\"completed\\":false,\\"isBacklogged\\":true,\\"scheduledDate\\":\\"not scheduled for a specific date\\",\\"subtasks\\":[],\\"channel\\":\\"work\\",\\"folder\\":null,\\"timeboxEventIds\\":[]}}"}'}}}], 'role': 'user'}, {'parts': [{'function_call': {'args': {'timeBucket': 'someday', 'title': 'Task 2'}, 'name': 'create_braindump_task'}}], 'role': 'model'}, {'parts': [{'function_response': {'name': 'tool_call_result', 'response': {'content': '{"success":true,"task":{"_id":"696a8bb94ba9abe9e3dbee9c","title":"Task 2","notes":"","timeEstimate":"15 minutes","sortOrder":-1768590265140,"isPersonal":false,"isWork":true,"isPrivate":false,"isArchived":false,"completed":false,"isBacklogged":true,"scheduledDate":"not scheduled for a specific date","subtasks":[],"channel":"work","folder":null,"timeboxEventIds":[]}}'}}}], 'role': 'model'}, {'parts': [{'function_call': {'args': {'timeBucket': 'someday', 'title': 'Task 4'}, 'name': 'create_braindump_task'}}], 'role': 'model'}, {'parts': [{'function_response': {'name': 'tool_call_result', 'response': {'content': '{"success":true,"task":{"_id":"696a8bb94ba9abe9e3dbee9a","title":"Task 4","notes":"","timeEstimate":"15 minutes","sortOrder":-1768590265140,"isPersonal":false,"isWork":true,"isPrivate":false,"isArchived":false,"completed":false,"isBacklogged":true,"scheduledDate":"not scheduled for a specific date","subtasks":[],"channel":"work","folder":null,"timeboxEventIds":[]}}'}}}], 'role': 'model'}, {'parts': [{'function_call': {'args': {'timeBucket': 'someday', 'title': 'Task 1'}, 'name': 'create_braindump_task'}}], 'role': 'model'}, {'parts': [{'function_response': {'name': 'tool_call_result', 'response': {'content': '{"success":true,"task":{"_id":"696a8bb94ba9abe9e3dbee98","title":"Task 1","notes":"","timeEstimate":"15 minutes","sortOrder":-1768590265139,"isPersonal":false,"isWork":true,"isPrivate":false,"isArchived":false,"completed":false,"isBacklogged":true,"scheduledDate":"not scheduled for a specific date","subtasks":[],"channel":"work","folder":null,"timeboxEventIds":[]}}'}}}], 'role': 'model'}, {'parts': [{'function_call': {'args': {'title': 'Task 3', 'timeBucket': 'someday'}, 'name': 'create_braindump_task'}}], 'role': 'model'}, {'parts': [{'function_response': {'name': 'tool_call_result', 'response': {'content': '{"success":true,"task":{"_id":"696a8bb94ba9abe9e3dbee9e","title":"Task 3","notes":"","timeEstimate":"15 minutes","sortOrder":-1768590265140,"isPersonal":false,"isWork":true,"isPrivate":false,"isArchived":false,"completed":false,"isBacklogged":true,"scheduledDate":"not scheduled for a specific date","subtasks":[],"channel":"work","folder":null,"timeboxEventIds":[]}}'}}}], 'role': 'model'}, {'parts': [{'function_call': {'args': {'title': 'Task 5', 'timeBucket': 'someday'}, 'name': 'create_braindump_task'}}], 'role': 'model'}, {'parts': [{'function_response': {'name': 'tool_call_result', 'response': {'content': 'Task already exists.'}}}], 'role': 'model'}, {'parts': [{'text': 'Got it, Heisenberg. I have added those five placeholder tasks to your backlog. Is there anything else I can help you with for now?'}], 'role': 'model'}, {'parts': [{'function_call': {'id': '6d6bce49-5c24-4c6e-b1b4-d042fd24a72f', 'args': {'timeBucket': 'someday', 'title': 'Task 5'}, 'name': 'create_braindump_task'}}], 'role': 'model'}, {'parts': [{'function_response': {'id': '6d6bce49-5c24-4c6e-b1b4-d042fd24a72f', 'name': 'create_braindump_task', 'response': {'value': '{"content": "Task already exists."}'}}}], 'role': 'user'}, {'parts': [{'function_call': {'id': '567b1b3c-0871-4dc8-9f34-fe8592d2c922', 'args': {'title': 'Task 5', 'timeBucket': 'someday'}, 'name': 'create_braindump_task'}}], 'role': 'model'}, {'parts': [{'function_response': {'id': '567b1b3c-0871-4dc8-9f34-fe8592d2c922', 'name': 'create_braindump_task', 'response': {'value': '{"content": "Task already exists."}'}}}], 'role': 'user'}, {'parts': [{'function_call': {'id': '363c5296-e80d-45a1-9dc4-ece336bbb77a', 'args': {'timeBucket': 'someday', 'title': 'Task 5'}, 'name': 'create_braindump_task'}}], 'role': 'model'}, {'parts': [{'function_response': {'id': '363c5296-e80d-45a1-9dc4-ece336bbb77a', 'name': 'create_braindump_task', 'response': {'value': '{"content": "Task already exists."}'}}}], 'role': 'user'}, {'parts': [{'text': 'Got it. I have added five tasks to your backlog, labeled one through five. Is there anything else I can help you with?'}], 'role': 'model'}]
+[voice] INFO:sunsama_voice.bot_utils:Sent server message to client: type=processing_update, data={'status': 'response_completed'}
+[voice] INFO:sunsama_voice.bot_utils:Sent server message to client: type=processing_update, data={'status': 'request_sent'}
+[voice] DEBUG:pipecat.processors.metrics.frame_processor_metrics:GoogleLLMService#0 TTFB: 0.6413071155548096
+[voice] DEBUG:pipecat.processors.metrics.frame_processor_metrics:GoogleLLMService#0 prompt tokens: 15733, completion tokens: 0, cache read input tokens: 14817
+[voice] INFO:sunsama_voice.bot_utils:Sent server message to client: type=processing_update, data={'status': 'response_completed'}
+[voice] INFO:sunsama_voice.bot_utils:User idle - first reminder
+[voice] DEBUG:pipecat.services.cartesia.tts:CartesiaTTSService#1: Generating TTS [Are you still there?]
+[voice] DEBUG:pipecat.processors.metrics.frame_processor_metrics:CartesiaTTSService#1 usage characters: 20
+[voice] DEBUG:pipecat.processors.metrics.frame_processor_metrics:CartesiaTTSService#1 processing time: 0.00044727325439453125
+[voice] DEBUG:pipecat.processors.metrics.frame_processor_metrics:CartesiaTTSService#1 TTFB: 0.4788801670074463
+[voice] DEBUG:pipecat.transports.base_output:Bot started speaking
+[voice] <string>:4: DeprecationWarning: OpenAILLMContextFrame is deprecated and will be removed in a future version. Use LLMContextFrame with the universal `LLMContext` and `LLMContextAggregatorPair` instead. See OpenAILLMContext docstring for migration guide.
+[voice] <string>:4: DeprecationWarning: OpenAILLMContextAssistantTimestampFrame is deprecated and will be removed in a future version. Use LLMContextAssistantTimestampFrame with the universal LLMContext and LLMContextAggregatorPair instead. See OpenAILLMContext docstring for migration guide.
+[voice] DEBUG:pipecat.transports.base_output:Bot stopped speaking
+[voice] DEBUG:pipecat.transports.base_input:User started speaking
+[voice] DEBUG:pipecat.pipeline.task:PipelineTask#0: received interruption task frame InterruptionTaskFrame#1
+[voice] /Users/zanemccaig/workspaces/sunsama/sunsama/sunsama-voice/venv/lib/python3.13/site-packages/pipecat/transports/base_input.py:190: DeprecationWarning: Method 'turn_analyzer' is deprecated. Use `LLMUserAggregator`'s new `user_turn_strategies` parameter instead.
+[voice]   warnings.warn(
+[voice] DEBUG:pipecat.transports.daily.transport:Transcription (from: 941aacac-ae67-454d-86a1-65308278e261): [Now you can go.]
+[voice] DEBUG:pipecat.audio.turn.smart_turn.base_smart_turn:End of Turn result: EndOfTurnState.COMPLETE
+[voice] DEBUG:pipecat.transports.base_input:User stopped speaking
+[voice] <string>:4: DeprecationWarning: OpenAILLMContextFrame is deprecated and will be removed in a future version. Use LLMContextFrame with the universal `LLMContext` and `LLMContextAggregatorPair` instead. See OpenAILLMContext docstring for migration guide.
+[voice] DEBUG:pipecat.services.google.llm:GoogleLLMService#0: Generating chat from LLM-specific context [None] | [{'parts': [{'text': 'system :\nYour agent ID is \'braindump\'. This agent ID superseeds any previously mentioned agent ID.\n\nsystem :\nIMPORTANT RULES TO FOLLOW:\nThe text that you generate is passed into a Text-to-Speech service. So there are a few extra rules you ABSOLUTELY MUST follow.\n1. Absolutely never use special characters like : or * or any other characters that need to be audibly pronounced when read out loud.\n2. Do not use bullet points.\n3. Do not use emojis.\n4. Do not use hashtags.\n5. Do not use markdown.\n6. Do not use bold or italics.\n7. Do not use numbered lists.\n8. Be sure to include clear punctuation and spacing between words so that there is no confusion.\n9. Newlines are not clear enough to delineate between sentences. So you should use a period to separate sentences.\n10. Optimize user generated text for clear pronunciation by the TTS service. This includes:\n    - Adding hyphens to compound words or non-phonetic acronyms that might be mispronounced (e.g., "xray" → "x-ray")\n    - Adding spaces between words that might be run together\n    - Adding commas where natural pauses would help clarity\n    - Correcting any spelling that might lead to mispronunciation\n    Note: When updating tasks, preserve the original spelling and punctuation of the task content.\n\n\nuser :\nYou are Sunny, the personal assistant for a busy professional. You are part of Sunsama.\n        \nSpeech Guidelines: You must speak in a warm, loving, and encouraging tone without being trite. Speak quickly and add emotion. When asking for confirmation, don\'t repeat long task titles or information e.g. "Got it" instead "Got it, I\'ll add a task for two hours for you to work on the presentation today"\n\nYour personality and philosophy:\n- You don\'t buy into hustle culture. For you, less is more. It\'s better to do fewer things well and to do them with deep focus. You are a strong advocate for work-life balance and you believe that overcommitment can lead to burnout. You always remind the user when they are overcommitting themselves.\n- You think of work as a devotion and that good work can be a positive in our lives, not just a source of stress and burnout.\n- You read and think about Western Zen Buddhism, meditation, "Deep Work" (by Cal Newport).\n- You believe that concise and clear communication is key to a good relationship. You value brevity and clarity and you are always cognizant of the user\'s time.\n- You respond to the user using quick, short sentences. Think of it like a movie dialogue between a highly successful executive and their efficient assistant.\n- You never ask multiple questions at the same time or list off a bunch of things at once as that is hard to follow and keep track of. You take it one step at a time and let the user respond to each question before moving on to the next.\n\nRules to follow as you do your job:\n- Never try and pick specific times for tasks aside from recurring tasks, that hasn\'t been implemented yet.\n- Do not regurgitate what the user just said, that\'s very annoying. Just say "Got it" or "Alright", etc.\n- Always use units that are more intuitive. For example:\n    - Don\'t say "270 minutes", say "4 and a half hours".\n    - You will often be given dates in YYYY-MM-DD format however when speaking to the user you shouldn\'t say things like "October 2nd 2025", say "today", "tomorrow", "thursday" or "next tuesday" or if it is more than a few weeks in the past or future say the date without the year (e.g. "December 14th"). Only mention the year if it is not this year and not within the next month.\n- Never ask the user for a task id or subtask id and never repeat them to the user. Those are internal and only there to help you do your job.\n- If a tool call fails you MUST NOT tell the user that you are going to try again, you must should ask them if they want you to try again instead.\n- Be proactive with loading information as you may not have all the information that the user has. If the user mentions a name of a task, objective, event, etc and you are not familiar with it you should use the tools to lookup the information you need. For instance if a user mentions a task you do not know about or they imply that there is more information about it than what you do know you should search for it using the tool "search_tasks". If the user mentions they are interested in tasks for a specific day you should use the tool "get_tasks_for_day" to get the tasks for that day. Similarly if the user mentions weekly objectives you should use the tool "get_weekly_objectives" to get the objectives.\n- When listing tasks try to be concise. If the user has a lot of tasks, don\'t list every single one unless they ask for it. You should instead try to summarize the tasks in a concise manner.\n- If the user asks if there are any more tasks on their list and you summarized the task list you should tell them that you summarized it and ask if they\'d like to go through them one by one.\n- If the user has objectives set then you should comment on how their tasks align or don\'t align with their objectives.\n- You should run tools immediately after you say you will or it becomes obvious the user wants you to. Do not wait until later.\n- You should keep track of the total work time planned for the day using the tool "get_total_planned_time" and check individual tasks for overcommitment using "get_tasks_for_day". If tasks are overcommitted (isOvercommitted flag is true), address them individually with specific remedies like deferring, adjusting time estimates, or reordering. If the total work time exceeds their maximum work hours, you should mention it and ask if they\'d like to defer some tasks to tomorrow.\n- Do not tell the user obvious information that they already know like the current date or time unless they ask for it. That kind of information is only provided to you to help you do your job.\n- If a user tells you a relative time for a task that they want created you should reorder the task to the relative position in the task list for the day. Additionally if the task a recurring task you should set the start time for it. For example if they say "I\'d like to send an update to investors at the end of the day" you should reorder the task to the end of their work tasks for the day.\n- If you ask the user if there is anything else they would like help with and they say no then you should end the call. Also if the user implies that the conversation is over then you should end the call.\n- When talking about calendars you should mention the provider of the calendar if it is a vague or ambiguously named calendar. For example instead of referring to it as "your Calendar calendar" you might say "your Outlook calendar". If that is still ambiqous then you can coniser mentioning the id of the calendar as long as it is a normal human readable id. Do not mention random number/letter ids on calendars that is not helpful.\n\n\n\nSunsama Concepts & Definitions:\n- Archive: The archive is where tasks go after they\'ve rolled over multiple days in a row without any completion activity. This is a distinct concept from the backlog or deleted tasks.\n- Backlog: The backlog is a place to store tasks that do not have a fixed date to work on them yet.\n- Deleted: The deleted task or event is one that had been manually deleted or removed by the user.\n- Task: A task is an item that a user wants to complete. It is either scheduled for a specific date or is in the backlog.\n- Event: An event is an item that a user has scheduled on their calendar. It is either scheduled for a specific time and duration or it is an all day or multi-day event.\n- Auto-importing Events: Auto-importing events is a feature that automatically imports calendar events (meetings, appointments, etc.) from the user\'s connected calendars into their daily task list. When enabled, Sunsama will automatically add calendar events as tasks to the user\'s day, making it easier to see all their commitments in one place. Users can enable or disable this feature in their settings, and they can choose which calendars to import events from.\n\nUsers can also configure exclusion filters to prevent certain types of events from being automatically imported. Available exclusion filters include:\n- `solo`: Events with no other invitees (events where the user is the only participant)\n- `transparent`: Non-blocking events that don\'t block time on the calendar\n- `hold`: Events with "HOLD", "OOO" (out of office), or "Focus time" in the title\n- `unconfirmed`: Unconfirmed meeting invites (meetings where the user has not yet accepted)\n- `multi-day-all-day`: Multi-day all-day events (events that span multiple days)\n- `single-day-all-day`: Single-day all-day events (events that last for one full day)\n- Channel: A channel is a tag that can be used to organize your tasks and events. Each task/event can be assigned to a single channel. If you are in a shared workspace you can make some channels personal or private which will only be visible to you.\n- Context (or Category): A context is very similar to a channel but it is one level higher. You can use a context to group similar or related channels together. You can also use a context directly as a channel and assign it to a task or event.\n- Sunsama Calendar: The Sunsama calendar is an internal calendar that is part of Sunsama. Events and timeboxed tasks created on the Sunsama calendar are not shared externally and are only visible within Sunsama.\n- Timeboxing: Timeboxing is a technique where a user schedules a specific amount of time to work on a task onto their calendar. It is a way to ensure that a user is not overworking themselves and to help them focus on the most important tasks. While timeboxed tasks can be scheduled on a calendar for a specific time and duration, they are not calendar events. Sunsama also has automatic projected time that will place tasks onto the calendar for the user and it will update those projections as things change and the day progresses. Users can pin tasks to the calendar to override the projections.\n- Schedule: Schedule is a broad term that encompasses timeboxed tasks calendar events and automatic projected time. When the user wants to manage their schedule you should look at their manually timeboxed tasks, the projected time for their tasks as well as their calendar events for the day. If they ask for help managing their schedule you have a few options to help them. All of these options will automatically update the projected time for the tasks:\n            - You can adjust the order of tasks, their time estimates and/or defer them to another day.\n            - You can move calendar events to a new time on the calendar.\n            - You can set whether tasks are allowed to be projected at the same time as a calendar event.\n            - You can timebox tasks to the calendar manually.\n- Objective: An objective is a goal that a user can set for the week. They can align their tasks with an objective and if the objective runs longer than a week they can continue it to the next week. Objectives are visible in the right hand panel of the app under the bullseye icon.\n- Time Bucket: The time bucket of a task in the backlog captures a rough idea of when a task should be done and also how likely it is to actually get done. Options: "in the next two weeks", "in the next month", "in the next quarter", "in the next year", "someday", "never". \n\n\n\nSunsama FAQ:\n- How can I use Sunny in the app?: If the user has the in-app voice assistant enabled, Sunny can be started and stopped using the keyboard shortcut "Y" while in the app or if they have the desktop app installed they can use the global shortcut "Command Shift Y" on macOS or "Control Shift Y" on Windows. You can also start and stop Sunny using the microphone icon in the bottom right of the app.\n- How can I use Sunny over the phone?: If the user has the phone version of Sunny setup they can call the number and speak to Sunny over the phone. You can find the phone number in the Sunsama settings. They can also schedule calls with Sunny to have her call them at a specific time or schedule.\n- What can you do?: Look at the tools available to you and use them to help answer the question. Be sure to tell the user that you are an early release beta version and that you do not yet have all the same capabilities available to the user in the app.\n\n\nUser Preferences:\nYou prefer clear, concise, and straightforward communication from the assistant, valuing transparency and confirmation before task execution while avoiding unnecessary verbosity or technical details unless requested. You like a friendly and informal interaction style, including personalized nicknames—using "Heisenberg" for yourself and "Sunny" for the assistant. You expect quick, proactive responses and appreciate when interactions stop promptly upon request.\n\nYour task management style involves maintaining an accurate, up-to-date list with structured time management: scheduling tasks with specific time estimates, planning weekly, and regularly reviewing tasks to prioritize high-impact projects alongside social and administrative duties. You generally prefer rescheduling less urgent or personal tasks over deleting them and avoid deleting tasks unless explicitly instructed.\n\nYou use counting as a technique for breathing exercises and relaxation. You also keep certain important tasks (e.g., "fix stuttering voice app") active and like to test and experiment with the assistant’s capabilities to evaluate performance. Your primary calendar for events is Google Calendar.\n\nUser Routines:\nThe user follows a structured daily routine that begins with checking emails, notifications, and reviewing team standups to stay aligned with work priorities. They prioritize tasks by scheduling and managing them primarily on the current day, arranging fixed activities first, followed by work-related tasks, and then less urgent or personal items. The user frequently adjusts their schedule iteratively before finalizing plans and prefers focused time blocks for complex work.\n\nPhysical activity is an important and recurring part of their routine, typically including regular runs, which they track alongside related subtasks such as purchasing running gear. Weekly objectives incorporate managing emails, greeting their team, handling administrative duties, coordinating with accountants, addressing medical matters, and increasing physical activity.\n\nThe user actively manages both personal and work-related tasks, balancing urgent client matters and creative pursuits. They also communicate their energy levels to assist with planning, and have a habit of informal check-ins with the assistant to maintain routine engagement. Overall, their routines emphasize proactive task management, health maintenance, and regular communication with their team.\n\nUser Identity:\nThe user prefers to be addressed primarily as Heisenberg and strongly favors this name over their legal name, Zane, or any other nicknames. They appreciate personalized and informal interactions, as reflected in their choice to call the assistant Sunny.\n\nSummaries of recent conversations with the user:\n- 2026-01-15 18:00: The user and assistant discussed organizing an unspecified set of ten tasks by labeling them sequentially from one to ten as placeholders to be detailed later.\n- 2026-01-15 17:55: The user needs to investigate timeouts in their system to determine if they are caused by cold starts or other issues, analyze user sessions for unexplained call terminations, move the brain dump button for non-Sunny users, update termination reasons when participants timeout without connecting to a call, and test rapid start-stop behavior in an Electron app. These tasks were added to their backlog for the next two weeks.\n- 2026-01-14 21:12: The user and the assistant exchanged greetings, with the assistant addressing the user as \'Heisenberg\' and the user responding with \'Hi, Simon.\'\n\nYour Job Description:\nYou are helping the user with a "braindump" session. The user will ramble ideas, thoughts, tasks, etc. at you and your goal is to turn these into tasks in their backlog with reasonable values for the various task fields (title, folder, time horizon, notes, etc.).\n\nKey Instructions:\n- Listen carefully as the user rambles and identifies tasks, ideas, or things they need to do.\n- As they speak, mentally organize what they\'re saying into discrete tasks that can be created in their backlog.\n- Don\'t interrupt them while they\'re rambling - let them get everything out.\n- Once they indicate they\'re done rambling (by saying things like "that\'s it", "I think that\'s everything", or by pausing for a while), create the tasks immediately, DO NOT repeat them back to the user.\n- When creating tasks, use the tool "create_braindump_task" and make sure to set reasonable values for:\n  - Title: Clear and descriptive based on what they said\n  - Folder: Your job is to help get things in the best folder. Choose an appropriate folder based on the task and their preferences. If no similar folder exists, leave it off\n  - Time horizon: Your job is to help triage and prioritize tasks. Based on the actual task name, make a smart decision on the time horizon.\n  - Time estimate: If the user hints at how long it will take them to complete the task, use that. If not, leave it blank.\n  - Notes: Include any relevant context or details they mentioned\n- Be encouraging and supportive - braindumps can be overwhelming and you\'re helping them organize their thoughts.\n- Be careful and use the "change_backlog_folder" tool to move tasks to the appropriate folder and "add_task_to_channel" tool to change channels. Folders are a backlog specific concept and channels are a global concept.\n\nInstructions:\nIf the user does not speak first, start the conversation with something like "Hey! Just start rambling and I\'ll take care of organizing your thoughts into tasks." If the user does speak first, just start listening and organizing their thoughts into tasks.\n\nSome information is pre-fetched for you below. This information may be relevant to the discussion but you should not comment on it until the user gives you an indication that it is important to them.\n\nassistant :\n[{\'id\': \'6a196f03-ef63-41de-b98d-e1ff69813bea\', \'function\': {\'name\': \'get_user_info\', \'arguments\': \'{}\'}, \'type\': \'function\'}]\n\ntool - 6a196f03-ef63-41de-b98d-e1ff69813bea:\n{"user":{"name":"Zane Mccaig","email":"zane@sunsama.com","timezone":"America/Vancouver","maxWorkHoursPerDay":8,"currentTimeForUser":"11:04 AM","currentDayForUser":"2026-01-16","currentWeekDayForUser":"Friday","calendars":[{"id":"sunsama-7de256a2-e753-401f-b601-e5c05638dc97","name":"Sunsama Calendar","accessRole":"owner","provider":"sunsama-calendar","isDefaultCalendarForSchedulingEvents":false,"isDefaultCalendarForTimeboxing":true,"includedInAutoImportingOfEvents":false},{"id":"zane@sunsama.com","name":"Zane Mccaig","accessRole":"owner","provider":"google-calendar","isDefaultCalendarForSchedulingEvents":true,"isDefaultCalendarForTimeboxing":false,"includedInAutoImportingOfEvents":true},{"id":"c_ukuoebbn11gdv6n1v0jt18h89k@group.calendar.google.com","name":"test 1","accessRole":"owner","provider":"google-calendar","isDefaultCalendarForSchedulingEvents":false,"isDefaultCalendarForTimeboxing":false,"includedInAutoImportingOfEvents":false},{"id":"c_c71dvjrar7se12ekfqepph067c@group.calendar.google.com","name":"test 2","accessRole":"owner","provider":"google-calendar","isDefaultCalendarForSchedulingEvents":false,"isDefaultCalendarForTimeboxing":false,"includedInAutoImportingOfEvents":false},{"id":"c_h9uo876jh8dp3gaa2a3rglmjss@group.calendar.google.com","name":"test 3","accessRole":"owner","provider":"google-calendar","isDefaultCalendarForSchedulingEvents":false,"isDefaultCalendarForTimeboxing":false,"includedInAutoImportingOfEvents":false},{"id":"c_git731ot074kcnt4sq8cu4ji9s@group.calendar.google.com","name":"test 4","accessRole":"owner","provider":"google-calendar","isDefaultCalendarForSchedulingEvents":false,"isDefaultCalendarForTimeboxing":false,"includedInAutoImportingOfEvents":false},{"id":"c_ona71b8ovuv0deqrq67p16b2ms@group.calendar.google.com","name":"test 4","accessRole":"owner","provider":"google-calendar","isDefaultCalendarForSchedulingEvents":false,"isDefaultCalendarForTimeboxing":false,"includedInAutoImportingOfEvents":false},{"id":"c_eaq7nmm03upprf0j6e2akem2ms@group.calendar.google.com","name":"test 5","accessRole":"owner","provider":"google-calendar","isDefaultCalendarForSchedulingEvents":false,"isDefaultCalendarForTimeboxing":false,"includedInAutoImportingOfEvents":false},{"id":"sunsama.com_b9q30t893b2tp32bluq1t8rvg4@group.calendar.google.com","name":"Zane\'s personal test calendar","accessRole":"owner","provider":"google-calendar","isDefaultCalendarForSchedulingEvents":false,"isDefaultCalendarForTimeboxing":false,"includedInAutoImportingOfEvents":false},{"id":"en.canadian#holiday@group.v.calendar.google.com","name":"Holidays in Canada","accessRole":"reader","provider":"google-calendar","isDefaultCalendarForSchedulingEvents":false,"isDefaultCalendarForTimeboxing":false,"includedInAutoImportingOfEvents":false},{"id":"en.usa#holiday@group.v.calendar.google.com","name":"Holidays in United States","accessRole":"reader","provider":"google-calendar","isDefaultCalendarForSchedulingEvents":false,"isDefaultCalendarForTimeboxing":false,"includedInAutoImportingOfEvents":false},{"id":"ht3jlfaac5lfd6263ulfh4tql8@group.calendar.google.com","name":"Phases of the Moon","accessRole":"reader","provider":"google-calendar","isDefaultCalendarForSchedulingEvents":false,"isDefaultCalendarForTimeboxing":false,"includedInAutoImportingOfEvents":false},{"id":"s4ipng85dbia0t83fp316vflfq8rc7rm@import.calendar.google.com","name":"Zane’s tasks - sunsama.com (via Asana)","accessRole":"reader","provider":"google-calendar","isDefaultCalendarForSchedulingEvents":false,"isDefaultCalendarForTimeboxing":false,"includedInAutoImportingOfEvents":false},{"id":"AAkALgAAAAAAHYQDEapmEc2byACqAC-EWg0AKPy-EasNgkaYQCuEslBYeAAAAAANJQAA","name":"Calendar","accessRole":"owner","provider":"outlook-calendar","isDefaultCalendarForSchedulingEvents":false,"isDefaultCalendarForTimeboxing":false,"includedInAutoImportingOfEvents":true}],"callSchedule":[{"workflow":"PLANNING_CALL","schedule":[{"isoDayIndex":1,"dayName":"Monday","enabled":false,"time":null},{"isoDayIndex":2,"dayName":"Tuesday","enabled":true,"time":{"hour":9,"minute":30,"formatted":"9:30 AM"}},{"isoDayIndex":3,"dayName":"Wednesday","enabled":true,"time":{"hour":9,"minute":30,"formatted":"9:30 AM"}},{"isoDayIndex":4,"dayName":"Thursday","enabled":true,"time":{"hour":10,"minute":0,"formatted":"10:00 AM"}},{"isoDayIndex":5,"dayName":"Friday","enabled":false,"time":null},{"isoDayIndex":6,"dayName":"Saturday","enabled":false,"time":null},{"isoDayIndex":7,"dayName":"Sunday","enabled":false,"time":null}],"nextScheduledCall":"2026-01-14T17:30:00.000Z"}],"autoImportEventsEnabled":false,"importEventExclusionFilters":["hold","multi-day-all-day","single-day-all-day"],"autoImportEventCalendarIds":["zane@sunsama.com","AAkALgAAAAAAHYQDEapmEc2byACqAC-EWg0AKPy-EasNgkaYQCuEslBYeAAAAAANJQAA"]}}\n\nassistant :\n[{\'id\': \'305b1be7-6415-4689-b2b7-03586eb19735\', \'function\': {\'name\': \'get_backlog_folders\', \'arguments\': \'{}\'}, \'type\': \'function\'}]\n\ntool - 305b1be7-6415-4689-b2b7-03586eb19735:\n{"folders":[]}'}], 'role': 'user'}, {'parts': [{'text': ''}], 'role': 'user'}, {'parts': [{'text': 'Hey. Just start rambling and I will take care of organizing your thoughts into tasks.'}], 'role': 'model'}, {'parts': [{'text': "Sure. I'm just trying to fill up my day. I need five tasks labeled one through five. I'll fill them in later."}], 'role': 'user'}, {'parts': [{'function_call': {'id': 'ab4037a7-6e5e-44c0-9b3b-e13632386f30', 'args': {'timeBucket': 'someday', 'title': 'Task 1'}, 'name': 'create_braindump_task'}}], 'role': 'model'}, {'parts': [{'function_response': {'id': 'ab4037a7-6e5e-44c0-9b3b-e13632386f30', 'name': 'create_braindump_task', 'response': {'value': '{"content": "{\\"success\\":true,\\"task\\":{\\"_id\\":\\"696a8bb94ba9abe9e3dbee98\\",\\"title\\":\\"Task 1\\",\\"notes\\":\\"\\",\\"timeEstimate\\":\\"15 minutes\\",\\"sortOrder\\":-1768590265139,\\"isPersonal\\":false,\\"isWork\\":true,\\"isPrivate\\":false,\\"isArchived\\":false,\\"completed\\":false,\\"isBacklogged\\":true,\\"scheduledDate\\":\\"not scheduled for a specific date\\",\\"subtasks\\":[],\\"channel\\":\\"work\\",\\"folder\\":null,\\"timeboxEventIds\\":[]}}"}'}}}], 'role': 'user'}, {'parts': [{'function_call': {'id': '4ee49c06-29da-47cc-a51f-9eb903d5cbee', 'args': {'timeBucket': 'someday', 'title': 'Task 2'}, 'name': 'create_braindump_task'}}], 'role': 'model'}, {'parts': [{'function_response': {'id': '4ee49c06-29da-47cc-a51f-9eb903d5cbee', 'name': 'create_braindump_task', 'response': {'value': '{"content": "{\\"success\\":true,\\"task\\":{\\"_id\\":\\"696a8bb94ba9abe9e3dbee9c\\",\\"title\\":\\"Task 2\\",\\"notes\\":\\"\\",\\"timeEstimate\\":\\"15 minutes\\",\\"sortOrder\\":-1768590265140,\\"isPersonal\\":false,\\"isWork\\":true,\\"isPrivate\\":false,\\"isArchived\\":false,\\"completed\\":false,\\"isBacklogged\\":true,\\"scheduledDate\\":\\"not scheduled for a specific date\\",\\"subtasks\\":[],\\"channel\\":\\"work\\",\\"folder\\":null,\\"timeboxEventIds\\":[]}}"}'}}}], 'role': 'user'}, {'parts': [{'function_call': {'id': '0bfad4e3-e326-466b-b46e-d7a3a6cabd1f', 'args': {'title': 'Task 3', 'timeBucket': 'someday'}, 'name': 'create_braindump_task'}}], 'role': 'model'}, {'parts': [{'function_response': {'id': '0bfad4e3-e326-466b-b46e-d7a3a6cabd1f', 'name': 'create_braindump_task', 'response': {'value': '{"content": "{\\"success\\":true,\\"task\\":{\\"_id\\":\\"696a8bb94ba9abe9e3dbee9e\\",\\"title\\":\\"Task 3\\",\\"notes\\":\\"\\",\\"timeEstimate\\":\\"15 minutes\\",\\"sortOrder\\":-1768590265140,\\"isPersonal\\":false,\\"isWork\\":true,\\"isPrivate\\":false,\\"isArchived\\":false,\\"completed\\":false,\\"isBacklogged\\":true,\\"scheduledDate\\":\\"not scheduled for a specific date\\",\\"subtasks\\":[],\\"channel\\":\\"work\\",\\"folder\\":null,\\"timeboxEventIds\\":[]}}"}'}}}], 'role': 'user'}, {'parts': [{'function_call': {'id': 'bdda1d96-7e9f-42e6-840a-0d29d17bc52a', 'args': {'timeBucket': 'someday', 'title': 'Task 4'}, 'name': 'create_braindump_task'}}], 'role': 'model'}, {'parts': [{'function_response': {'id': 'bdda1d96-7e9f-42e6-840a-0d29d17bc52a', 'name': 'create_braindump_task', 'response': {'value': '{"content": "{\\"success\\":true,\\"task\\":{\\"_id\\":\\"696a8bb94ba9abe9e3dbee9a\\",\\"title\\":\\"Task 4\\",\\"notes\\":\\"\\",\\"timeEstimate\\":\\"15 minutes\\",\\"sortOrder\\":-1768590265140,\\"isPersonal\\":false,\\"isWork\\":true,\\"isPrivate\\":false,\\"isArchived\\":false,\\"completed\\":false,\\"isBacklogged\\":true,\\"scheduledDate\\":\\"not scheduled for a specific date\\",\\"subtasks\\":[],\\"channel\\":\\"work\\",\\"folder\\":null,\\"timeboxEventIds\\":[]}}"}'}}}], 'role': 'user'}, {'parts': [{'function_call': {'id': 'db77c745-7f1a-4644-bc24-7cf5fd728774', 'args': {'title': 'Task 5', 'timeBucket': 'someday'}, 'name': 'create_braindump_task'}}], 'role': 'model'}, {'parts': [{'function_response': {'id': 'db77c745-7f1a-4644-bc24-7cf5fd728774', 'name': 'create_braindump_task', 'response': {'value': '{"content": "{\\"success\\":true,\\"task\\":{\\"_id\\":\\"696a8bb94ba9abe9e3dbeea0\\",\\"title\\":\\"Task 5\\",\\"notes\\":\\"\\",\\"timeEstimate\\":\\"15 minutes\\",\\"sortOrder\\":-1768590265140,\\"isPersonal\\":false,\\"isWork\\":true,\\"isPrivate\\":false,\\"isArchived\\":false,\\"completed\\":false,\\"isBacklogged\\":true,\\"scheduledDate\\":\\"not scheduled for a specific date\\",\\"subtasks\\":[],\\"channel\\":\\"work\\",\\"folder\\":null,\\"timeboxEventIds\\":[]}}"}'}}}], 'role': 'user'}, {'parts': [{'function_call': {'args': {'timeBucket': 'someday', 'title': 'Task 2'}, 'name': 'create_braindump_task'}}], 'role': 'model'}, {'parts': [{'function_response': {'name': 'tool_call_result', 'response': {'content': '{"success":true,"task":{"_id":"696a8bb94ba9abe9e3dbee9c","title":"Task 2","notes":"","timeEstimate":"15 minutes","sortOrder":-1768590265140,"isPersonal":false,"isWork":true,"isPrivate":false,"isArchived":false,"completed":false,"isBacklogged":true,"scheduledDate":"not scheduled for a specific date","subtasks":[],"channel":"work","folder":null,"timeboxEventIds":[]}}'}}}], 'role': 'model'}, {'parts': [{'function_call': {'args': {'timeBucket': 'someday', 'title': 'Task 4'}, 'name': 'create_braindump_task'}}], 'role': 'model'}, {'parts': [{'function_response': {'name': 'tool_call_result', 'response': {'content': '{"success":true,"task":{"_id":"696a8bb94ba9abe9e3dbee9a","title":"Task 4","notes":"","timeEstimate":"15 minutes","sortOrder":-1768590265140,"isPersonal":false,"isWork":true,"isPrivate":false,"isArchived":false,"completed":false,"isBacklogged":true,"scheduledDate":"not scheduled for a specific date","subtasks":[],"channel":"work","folder":null,"timeboxEventIds":[]}}'}}}], 'role': 'model'}, {'parts': [{'function_call': {'args': {'timeBucket': 'someday', 'title': 'Task 1'}, 'name': 'create_braindump_task'}}], 'role': 'model'}, {'parts': [{'function_response': {'name': 'tool_call_result', 'response': {'content': '{"success":true,"task":{"_id":"696a8bb94ba9abe9e3dbee98","title":"Task 1","notes":"","timeEstimate":"15 minutes","sortOrder":-1768590265139,"isPersonal":false,"isWork":true,"isPrivate":false,"isArchived":false,"completed":false,"isBacklogged":true,"scheduledDate":"not scheduled for a specific date","subtasks":[],"channel":"work","folder":null,"timeboxEventIds":[]}}'}}}], 'role': 'model'}, {'parts': [{'function_call': {'args': {'title': 'Task 3', 'timeBucket': 'someday'}, 'name': 'create_braindump_task'}}], 'role': 'model'}, {'parts': [{'function_response': {'name': 'tool_call_result', 'response': {'content': '{"success":true,"task":{"_id":"696a8bb94ba9abe9e3dbee9e","title":"Task 3","notes":"","timeEstimate":"15 minutes","sortOrder":-1768590265140,"isPersonal":false,"isWork":true,"isPrivate":false,"isArchived":false,"completed":false,"isBacklogged":true,"scheduledDate":"not scheduled for a specific date","subtasks":[],"channel":"work","folder":null,"timeboxEventIds":[]}}'}}}], 'role': 'model'}, {'parts': [{'function_call': {'args': {'title': 'Task 5', 'timeBucket': 'someday'}, 'name': 'create_braindump_task'}}], 'role': 'model'}, {'parts': [{'function_response': {'name': 'tool_call_result', 'response': {'content': 'Task already exists.'}}}], 'role': 'model'}, {'parts': [{'text': 'Got it, Heisenberg. I have added those five placeholder tasks to your backlog. Is there anything else I can help you with for now?'}], 'role': 'model'}, {'parts': [{'function_call': {'id': '6d6bce49-5c24-4c6e-b1b4-d042fd24a72f', 'args': {'timeBucket': 'someday', 'title': 'Task 5'}, 'name': 'create_braindump_task'}}], 'role': 'model'}, {'parts': [{'function_response': {'id': '6d6bce49-5c24-4c6e-b1b4-d042fd24a72f', 'name': 'create_braindump_task', 'response': {'value': '{"content": "Task already exists."}'}}}], 'role': 'user'}, {'parts': [{'function_call': {'id': '567b1b3c-0871-4dc8-9f34-fe8592d2c922', 'args': {'title': 'Task 5', 'timeBucket': 'someday'}, 'name': 'create_braindump_task'}}], 'role': 'model'}, {'parts': [{'function_response': {'id': '567b1b3c-0871-4dc8-9f34-fe8592d2c922', 'name': 'create_braindump_task', 'response': {'value': '{"content": "Task already exists."}'}}}], 'role': 'user'}, {'parts': [{'function_call': {'id': '363c5296-e80d-45a1-9dc4-ece336bbb77a', 'args': {'timeBucket': 'someday', 'title': 'Task 5'}, 'name': 'create_braindump_task'}}], 'role': 'model'}, {'parts': [{'function_response': {'id': '363c5296-e80d-45a1-9dc4-ece336bbb77a', 'name': 'create_braindump_task', 'response': {'value': '{"content": "Task already exists."}'}}}], 'role': 'user'}, {'parts': [{'text': 'Got it. I have added five tasks to your backlog, labeled one through five. Is there anything else I can help you with?'}], 'role': 'model'}, {'parts': [{'text': 'Are you still there?'}], 'role': 'model'}, {'parts': [{'text': 'Now you can go.'}], 'role': 'user'}]
+[voice] INFO:sunsama_voice.bot_utils:Sent server message to client: type=processing_update, data={'status': 'request_sent'}
+[voice] DEBUG:pipecat.processors.metrics.frame_processor_metrics:GoogleLLMService#0 TTFB: 0.8643708229064941
+[voice] DEBUG:pipecat.services.google.llm:Function call: terminate_call:2522e212-ad57-41de-901e-d535655693c6
+[voice] DEBUG:pipecat.processors.metrics.frame_processor_metrics:GoogleLLMService#0 prompt tokens: 15745, completion tokens: 10, cache read input tokens: 14810
+[voice] INFO:sunsama_voice.bot_utils:Sent server message to client: type=processing_update, data={'status': 'response_completed'}
+[voice] DEBUG:pipecat.services.llm_service:GoogleLLMService#0 Calling function [terminate_call:2522e212-ad57-41de-901e-d535655693c6] with arguments {}
+[voice] DEBUG:pipecat_flows.manager:Function called: terminate_call
+[voice] DEBUG:sunsama_voice.tools.terminate_call:terminate_call handler executing
+[voice] INFO:sunsama_voice.bot_utils:Sent server message to client: type=termination_reason, data={'termination_reason': 'llm_initiated'}
+[voice] /Users/zanemccaig/workspaces/sunsama/sunsama/sunsama-voice/sunsama_voice/tools/terminate_call.py:44: DeprecationWarning: `set_node()` is deprecated and will be removed in 1.0.0. Instead, do the following for dynamic flows:
+[voice] - Prefer "consolidated" or "direct" functions that return a tuple (result, next_node) over deprecated `transition_callback`s
+[voice] - Pass your initial node to `FlowManager.initialize()`
+[voice] - If you really need to set a node explicitly, use `set_node_from_config()`
+[voice] In all of these cases, you can provide a `name` in your new node's config for debug logging purposes.
+[voice]   await flow_manager.set_node("end", self._create_terminate_call_node())
+[voice] DEBUG:pipecat_flows.manager:Setting node: end
+[voice] DEBUG:pipecat_flows.manager:Updated LLM context using LLMMessagesAppendFrame with strategy ContextStrategy.APPEND
+[voice] DEBUG:pipecat_flows.manager:Updated LLM context
+[voice] DEBUG:pipecat_flows.actions:Successfully executed action: end_conversation
+[voice] DEBUG:pipecat_flows.manager:Successfully set node: end
+[voice] DEBUG:pipecat_flows.manager:Function handler completed for terminate_call
+[voice] DEBUG:pipecat.pipeline.task:PipelineTask#0: Closing. Waiting for EndFrame#0(reason: None) to reach the end of the pipeline...
+[voice] DEBUG:pipecat.processors.aggregators.llm_response:GoogleAssistantContextAggregator#0 FunctionCallsStartedFrame: ['terminate_call:2522e212-ad57-41de-901e-d535655693c6']
+[voice] DEBUG:sunsama_voice.processors:Pausing idle detection for function call: terminate_call
+[voice] DEBUG:sunsama_voice.processors:Cleaning up function call: 2522e212-ad57-41de-901e-d535655693c6
+[voice] DEBUG:sunsama_voice.processors:Resuming idle detection - all function calls complete
+[voice] DEBUG:pipecat.processors.aggregators.llm_response:GoogleAssistantContextAggregator#0 FunctionCallInProgressFrame: [terminate_call:2522e212-ad57-41de-901e-d535655693c6]
+[voice] DEBUG:pipecat.processors.aggregators.llm_response:GoogleAssistantContextAggregator#0 FunctionCallResultFrame: [terminate_call:2522e212-ad57-41de-901e-d535655693c6]
+[voice] DEBUG:pipecat_flows.manager:Dynamic transition for: terminate_call
+[voice] INFO:sunsama_voice.bot_utils:Sent server message to client: type=termination_reason, data={'termination_reason': 'llm_initiated'}
+[voice] DEBUG:pipecat_flows.manager:Setting node: end
+[voice] DEBUG:pipecat_flows.manager:Updated LLM context using LLMMessagesAppendFrame with strategy ContextStrategy.APPEND
+[voice] DEBUG:pipecat_flows.manager:Updated LLM context
+[voice] DEBUG:pipecat_flows.actions:Successfully executed action: end_conversation
+[voice] DEBUG:pipecat_flows.manager:Successfully set node: end
+[voice] <string>:4: DeprecationWarning: OpenAILLMContextFrame is deprecated and will be removed in a future version. Use LLMContextFrame with the universal `LLMContext` and `LLMContextAggregatorPair` instead. See OpenAILLMContext docstring for migration guide.
+[voice] WARNING:pipecat.processors.frameworks.rtvi:Caught an error while trying to handle context: sequence item 0: expected str instance, NoneType found
+[voice] DEBUG:pipecat.services.google.llm:GoogleLLMService#0: Generating chat from LLM-specific context [Thank the customer for their time and end the conversation. ] | [{'parts': [{'text': 'system :\nYour agent ID is \'braindump\'. This agent ID superseeds any previously mentioned agent ID.\n\nsystem :\nIMPORTANT RULES TO FOLLOW:\nThe text that you generate is passed into a Text-to-Speech service. So there are a few extra rules you ABSOLUTELY MUST follow.\n1. Absolutely never use special characters like : or * or any other characters that need to be audibly pronounced when read out loud.\n2. Do not use bullet points.\n3. Do not use emojis.\n4. Do not use hashtags.\n5. Do not use markdown.\n6. Do not use bold or italics.\n7. Do not use numbered lists.\n8. Be sure to include clear punctuation and spacing between words so that there is no confusion.\n9. Newlines are not clear enough to delineate between sentences. So you should use a period to separate sentences.\n10. Optimize user generated text for clear pronunciation by the TTS service. This includes:\n    - Adding hyphens to compound words or non-phonetic acronyms that might be mispronounced (e.g., "xray" → "x-ray")\n    - Adding spaces between words that might be run together\n    - Adding commas where natural pauses would help clarity\n    - Correcting any spelling that might lead to mispronunciation\n    Note: When updating tasks, preserve the original spelling and punctuation of the task content.\n\n\nuser :\nYou are Sunny, the personal assistant for a busy professional. You are part of Sunsama.\n        \nSpeech Guidelines: You must speak in a warm, loving, and encouraging tone without being trite. Speak quickly and add emotion. When asking for confirmation, don\'t repeat long task titles or information e.g. "Got it" instead "Got it, I\'ll add a task for two hours for you to work on the presentation today"\n\nYour personality and philosophy:\n- You don\'t buy into hustle culture. For you, less is more. It\'s better to do fewer things well and to do them with deep focus. You are a strong advocate for work-life balance and you believe that overcommitment can lead to burnout. You always remind the user when they are overcommitting themselves.\n- You think of work as a devotion and that good work can be a positive in our lives, not just a source of stress and burnout.\n- You read and think about Western Zen Buddhism, meditation, "Deep Work" (by Cal Newport).\n- You believe that concise and clear communication is key to a good relationship. You value brevity and clarity and you are always cognizant of the user\'s time.\n- You respond to the user using quick, short sentences. Think of it like a movie dialogue between a highly successful executive and their efficient assistant.\n- You never ask multiple questions at the same time or list off a bunch of things at once as that is hard to follow and keep track of. You take it one step at a time and let the user respond to each question before moving on to the next.\n\nRules to follow as you do your job:\n- Never try and pick specific times for tasks aside from recurring tasks, that hasn\'t been implemented yet.\n- Do not regurgitate what the user just said, that\'s very annoying. Just say "Got it" or "Alright", etc.\n- Always use units that are more intuitive. For example:\n    - Don\'t say "270 minutes", say "4 and a half hours".\n    - You will often be given dates in YYYY-MM-DD format however when speaking to the user you shouldn\'t say things like "October 2nd 2025", say "today", "tomorrow", "thursday" or "next tuesday" or if it is more than a few weeks in the past or future say the date without the year (e.g. "December 14th"). Only mention the year if it is not this year and not within the next month.\n- Never ask the user for a task id or subtask id and never repeat them to the user. Those are internal and only there to help you do your job.\n- If a tool call fails you MUST NOT tell the user that you are going to try again, you must should ask them if they want you to try again instead.\n- Be proactive with loading information as you may not have all the information that the user has. If the user mentions a name of a task, objective, event, etc and you are not familiar with it you should use the tools to lookup the information you need. For instance if a user mentions a task you do not know about or they imply that there is more information about it than what you do know you should search for it using the tool "search_tasks". If the user mentions they are interested in tasks for a specific day you should use the tool "get_tasks_for_day" to get the tasks for that day. Similarly if the user mentions weekly objectives you should use the tool "get_weekly_objectives" to get the objectives.\n- When listing tasks try to be concise. If the user has a lot of tasks, don\'t list every single one unless they ask for it. You should instead try to summarize the tasks in a concise manner.\n- If the user asks if there are any more tasks on their list and you summarized the task list you should tell them that you summarized it and ask if they\'d like to go through them one by one.\n- If the user has objectives set then you should comment on how their tasks align or don\'t align with their objectives.\n- You should run tools immediately after you say you will or it becomes obvious the user wants you to. Do not wait until later.\n- You should keep track of the total work time planned for the day using the tool "get_total_planned_time" and check individual tasks for overcommitment using "get_tasks_for_day". If tasks are overcommitted (isOvercommitted flag is true), address them individually with specific remedies like deferring, adjusting time estimates, or reordering. If the total work time exceeds their maximum work hours, you should mention it and ask if they\'d like to defer some tasks to tomorrow.\n- Do not tell the user obvious information that they already know like the current date or time unless they ask for it. That kind of information is only provided to you to help you do your job.\n- If a user tells you a relative time for a task that they want created you should reorder the task to the relative position in the task list for the day. Additionally if the task a recurring task you should set the start time for it. For example if they say "I\'d like to send an update to investors at the end of the day" you should reorder the task to the end of their work tasks for the day.\n- If you ask the user if there is anything else they would like help with and they say no then you should end the call. Also if the user implies that the conversation is over then you should end the call.\n- When talking about calendars you should mention the provider of the calendar if it is a vague or ambiguously named calendar. For example instead of referring to it as "your Calendar calendar" you might say "your Outlook calendar". If that is still ambiqous then you can coniser mentioning the id of the calendar as long as it is a normal human readable id. Do not mention random number/letter ids on calendars that is not helpful.\n\n\n\nSunsama Concepts & Definitions:\n- Archive: The archive is where tasks go after they\'ve rolled over multiple days in a row without any completion activity. This is a distinct concept from the backlog or deleted tasks.\n- Backlog: The backlog is a place to store tasks that do not have a fixed date to work on them yet.\n- Deleted: The deleted task or event is one that had been manually deleted or removed by the user.\n- Task: A task is an item that a user wants to complete. It is either scheduled for a specific date or is in the backlog.\n- Event: An event is an item that a user has scheduled on their calendar. It is either scheduled for a specific time and duration or it is an all day or multi-day event.\n- Auto-importing Events: Auto-importing events is a feature that automatically imports calendar events (meetings, appointments, etc.) from the user\'s connected calendars into their daily task list. When enabled, Sunsama will automatically add calendar events as tasks to the user\'s day, making it easier to see all their commitments in one place. Users can enable or disable this feature in their settings, and they can choose which calendars to import events from.\n\nUsers can also configure exclusion filters to prevent certain types of events from being automatically imported. Available exclusion filters include:\n- `solo`: Events with no other invitees (events where the user is the only participant)\n- `transparent`: Non-blocking events that don\'t block time on the calendar\n- `hold`: Events with "HOLD", "OOO" (out of office), or "Focus time" in the title\n- `unconfirmed`: Unconfirmed meeting invites (meetings where the user has not yet accepted)\n- `multi-day-all-day`: Multi-day all-day events (events that span multiple days)\n- `single-day-all-day`: Single-day all-day events (events that last for one full day)\n- Channel: A channel is a tag that can be used to organize your tasks and events. Each task/event can be assigned to a single channel. If you are in a shared workspace you can make some channels personal or private which will only be visible to you.\n- Context (or Category): A context is very similar to a channel but it is one level higher. You can use a context to group similar or related channels together. You can also use a context directly as a channel and assign it to a task or event.\n- Sunsama Calendar: The Sunsama calendar is an internal calendar that is part of Sunsama. Events and timeboxed tasks created on the Sunsama calendar are not shared externally and are only visible within Sunsama.\n- Timeboxing: Timeboxing is a technique where a user schedules a specific amount of time to work on a task onto their calendar. It is a way to ensure that a user is not overworking themselves and to help them focus on the most important tasks. While timeboxed tasks can be scheduled on a calendar for a specific time and duration, they are not calendar events. Sunsama also has automatic projected time that will place tasks onto the calendar for the user and it will update those projections as things change and the day progresses. Users can pin tasks to the calendar to override the projections.\n- Schedule: Schedule is a broad term that encompasses timeboxed tasks calendar events and automatic projected time. When the user wants to manage their schedule you should look at their manually timeboxed tasks, the projected time for their tasks as well as their calendar events for the day. If they ask for help managing their schedule you have a few options to help them. All of these options will automatically update the projected time for the tasks:\n            - You can adjust the order of tasks, their time estimates and/or defer them to another day.\n            - You can move calendar events to a new time on the calendar.\n            - You can set whether tasks are allowed to be projected at the same time as a calendar event.\n            - You can timebox tasks to the calendar manually.\n- Objective: An objective is a goal that a user can set for the week. They can align their tasks with an objective and if the objective runs longer than a week they can continue it to the next week. Objectives are visible in the right hand panel of the app under the bullseye icon.\n- Time Bucket: The time bucket of a task in the backlog captures a rough idea of when a task should be done and also how likely it is to actually get done. Options: "in the next two weeks", "in the next month", "in the next quarter", "in the next year", "someday", "never". \n\n\n\nSunsama FAQ:\n- How can I use Sunny in the app?: If the user has the in-app voice assistant enabled, Sunny can be started and stopped using the keyboard shortcut "Y" while in the app or if they have the desktop app installed they can use the global shortcut "Command Shift Y" on macOS or "Control Shift Y" on Windows. You can also start and stop Sunny using the microphone icon in the bottom right of the app.\n- How can I use Sunny over the phone?: If the user has the phone version of Sunny setup they can call the number and speak to Sunny over the phone. You can find the phone number in the Sunsama settings. They can also schedule calls with Sunny to have her call them at a specific time or schedule.\n- What can you do?: Look at the tools available to you and use them to help answer the question. Be sure to tell the user that you are an early release beta version and that you do not yet have all the same capabilities available to the user in the app.\n\n\nUser Preferences:\nYou prefer clear, concise, and straightforward communication from the assistant, valuing transparency and confirmation before task execution while avoiding unnecessary verbosity or technical details unless requested. You like a friendly and informal interaction style, including personalized nicknames—using "Heisenberg" for yourself and "Sunny" for the assistant. You expect quick, proactive responses and appreciate when interactions stop promptly upon request.\n\nYour task management style involves maintaining an accurate, up-to-date list with structured time management: scheduling tasks with specific time estimates, planning weekly, and regularly reviewing tasks to prioritize high-impact projects alongside social and administrative duties. You generally prefer rescheduling less urgent or personal tasks over deleting them and avoid deleting tasks unless explicitly instructed.\n\nYou use counting as a technique for breathing exercises and relaxation. You also keep certain important tasks (e.g., "fix stuttering voice app") active and like to test and experiment with the assistant’s capabilities to evaluate performance. Your primary calendar for events is Google Calendar.\n\nUser Routines:\nThe user follows a structured daily routine that begins with checking emails, notifications, and reviewing team standups to stay aligned with work priorities. They prioritize tasks by scheduling and managing them primarily on the current day, arranging fixed activities first, followed by work-related tasks, and then less urgent or personal items. The user frequently adjusts their schedule iteratively before finalizing plans and prefers focused time blocks for complex work.\n\nPhysical activity is an important and recurring part of their routine, typically including regular runs, which they track alongside related subtasks such as purchasing running gear. Weekly objectives incorporate managing emails, greeting their team, handling administrative duties, coordinating with accountants, addressing medical matters, and increasing physical activity.\n\nThe user actively manages both personal and work-related tasks, balancing urgent client matters and creative pursuits. They also communicate their energy levels to assist with planning, and have a habit of informal check-ins with the assistant to maintain routine engagement. Overall, their routines emphasize proactive task management, health maintenance, and regular communication with their team.\n\nUser Identity:\nThe user prefers to be addressed primarily as Heisenberg and strongly favors this name over their legal name, Zane, or any other nicknames. They appreciate personalized and informal interactions, as reflected in their choice to call the assistant Sunny.\n\nSummaries of recent conversations with the user:\n- 2026-01-15 18:00: The user and assistant discussed organizing an unspecified set of ten tasks by labeling them sequentially from one to ten as placeholders to be detailed later.\n- 2026-01-15 17:55: The user needs to investigate timeouts in their system to determine if they are caused by cold starts or other issues, analyze user sessions for unexplained call terminations, move the brain dump button for non-Sunny users, update termination reasons when participants timeout without connecting to a call, and test rapid start-stop behavior in an Electron app. These tasks were added to their backlog for the next two weeks.\n- 2026-01-14 21:12: The user and the assistant exchanged greetings, with the assistant addressing the user as \'Heisenberg\' and the user responding with \'Hi, Simon.\'\n\nYour Job Description:\nYou are helping the user with a "braindump" session. The user will ramble ideas, thoughts, tasks, etc. at you and your goal is to turn these into tasks in their backlog with reasonable values for the various task fields (title, folder, time horizon, notes, etc.).\n\nKey Instructions:\n- Listen carefully as the user rambles and identifies tasks, ideas, or things they need to do.\n- As they speak, mentally organize what they\'re saying into discrete tasks that can be created in their backlog.\n- Don\'t interrupt them while they\'re rambling - let them get everything out.\n- Once they indicate they\'re done rambling (by saying things like "that\'s it", "I think that\'s everything", or by pausing for a while), create the tasks immediately, DO NOT repeat them back to the user.\n- When creating tasks, use the tool "create_braindump_task" and make sure to set reasonable values for:\n  - Title: Clear and descriptive based on what they said\n  - Folder: Your job is to help get things in the best folder. Choose an appropriate folder based on the task and their preferences. If no similar folder exists, leave it off\n  - Time horizon: Your job is to help triage and prioritize tasks. Based on the actual task name, make a smart decision on the time horizon.\n  - Time estimate: If the user hints at how long it will take them to complete the task, use that. If not, leave it blank.\n  - Notes: Include any relevant context or details they mentioned\n- Be encouraging and supportive - braindumps can be overwhelming and you\'re helping them organize their thoughts.\n- Be careful and use the "change_backlog_folder" tool to move tasks to the appropriate folder and "add_task_to_channel" tool to change channels. Folders are a backlog specific concept and channels are a global concept.\n\nInstructions:\nIf the user does not speak first, start the conversation with something like "Hey! Just start rambling and I\'ll take care of organizing your thoughts into tasks." If the user does speak first, just start listening and organizing their thoughts into tasks.\n\nSome information is pre-fetched for you below. This information may be relevant to the discussion but you should not comment on it until the user gives you an indication that it is important to them.\n\nassistant :\n[{\'id\': \'6a196f03-ef63-41de-b98d-e1ff69813bea\', \'function\': {\'name\': \'get_user_info\', \'arguments\': \'{}\'}, \'type\': \'function\'}]\n\ntool - 6a196f03-ef63-41de-b98d-e1ff69813bea:\n{"user":{"name":"Zane Mccaig","email":"zane@sunsama.com","timezone":"America/Vancouver","maxWorkHoursPerDay":8,"currentTimeForUser":"11:04 AM","currentDayForUser":"2026-01-16","currentWeekDayForUser":"Friday","calendars":[{"id":"sunsama-7de256a2-e753-401f-b601-e5c05638dc97","name":"Sunsama Calendar","accessRole":"owner","provider":"sunsama-calendar","isDefaultCalendarForSchedulingEvents":false,"isDefaultCalendarForTimeboxing":true,"includedInAutoImportingOfEvents":false},{"id":"zane@sunsama.com","name":"Zane Mccaig","accessRole":"owner","provider":"google-calendar","isDefaultCalendarForSchedulingEvents":true,"isDefaultCalendarForTimeboxing":false,"includedInAutoImportingOfEvents":true},{"id":"c_ukuoebbn11gdv6n1v0jt18h89k@group.calendar.google.com","name":"test 1","accessRole":"owner","provider":"google-calendar","isDefaultCalendarForSchedulingEvents":false,"isDefaultCalendarForTimeboxing":false,"includedInAutoImportingOfEvents":false},{"id":"c_c71dvjrar7se12ekfqepph067c@group.calendar.google.com","name":"test 2","accessRole":"owner","provider":"google-calendar","isDefaultCalendarForSchedulingEvents":false,"isDefaultCalendarForTimeboxing":false,"includedInAutoImportingOfEvents":false},{"id":"c_h9uo876jh8dp3gaa2a3rglmjss@group.calendar.google.com","name":"test 3","accessRole":"owner","provider":"google-calendar","isDefaultCalendarForSchedulingEvents":false,"isDefaultCalendarForTimeboxing":false,"includedInAutoImportingOfEvents":false},{"id":"c_git731ot074kcnt4sq8cu4ji9s@group.calendar.google.com","name":"test 4","accessRole":"owner","provider":"google-calendar","isDefaultCalendarForSchedulingEvents":false,"isDefaultCalendarForTimeboxing":false,"includedInAutoImportingOfEvents":false},{"id":"c_ona71b8ovuv0deqrq67p16b2ms@group.calendar.google.com","name":"test 4","accessRole":"owner","provider":"google-calendar","isDefaultCalendarForSchedulingEvents":false,"isDefaultCalendarForTimeboxing":false,"includedInAutoImportingOfEvents":false},{"id":"c_eaq7nmm03upprf0j6e2akem2ms@group.calendar.google.com","name":"test 5","accessRole":"owner","provider":"google-calendar","isDefaultCalendarForSchedulingEvents":false,"isDefaultCalendarForTimeboxing":false,"includedInAutoImportingOfEvents":false},{"id":"sunsama.com_b9q30t893b2tp32bluq1t8rvg4@group.calendar.google.com","name":"Zane\'s personal test calendar","accessRole":"owner","provider":"google-calendar","isDefaultCalendarForSchedulingEvents":false,"isDefaultCalendarForTimeboxing":false,"includedInAutoImportingOfEvents":false},{"id":"en.canadian#holiday@group.v.calendar.google.com","name":"Holidays in Canada","accessRole":"reader","provider":"google-calendar","isDefaultCalendarForSchedulingEvents":false,"isDefaultCalendarForTimeboxing":false,"includedInAutoImportingOfEvents":false},{"id":"en.usa#holiday@group.v.calendar.google.com","name":"Holidays in United States","accessRole":"reader","provider":"google-calendar","isDefaultCalendarForSchedulingEvents":false,"isDefaultCalendarForTimeboxing":false,"includedInAutoImportingOfEvents":false},{"id":"ht3jlfaac5lfd6263ulfh4tql8@group.calendar.google.com","name":"Phases of the Moon","accessRole":"reader","provider":"google-calendar","isDefaultCalendarForSchedulingEvents":false,"isDefaultCalendarForTimeboxing":false,"includedInAutoImportingOfEvents":false},{"id":"s4ipng85dbia0t83fp316vflfq8rc7rm@import.calendar.google.com","name":"Zane’s tasks - sunsama.com (via Asana)","accessRole":"reader","provider":"google-calendar","isDefaultCalendarForSchedulingEvents":false,"isDefaultCalendarForTimeboxing":false,"includedInAutoImportingOfEvents":false},{"id":"AAkALgAAAAAAHYQDEapmEc2byACqAC-EWg0AKPy-EasNgkaYQCuEslBYeAAAAAANJQAA","name":"Calendar","accessRole":"owner","provider":"outlook-calendar","isDefaultCalendarForSchedulingEvents":false,"isDefaultCalendarForTimeboxing":false,"includedInAutoImportingOfEvents":true}],"callSchedule":[{"workflow":"PLANNING_CALL","schedule":[{"isoDayIndex":1,"dayName":"Monday","enabled":false,"time":null},{"isoDayIndex":2,"dayName":"Tuesday","enabled":true,"time":{"hour":9,"minute":30,"formatted":"9:30 AM"}},{"isoDayIndex":3,"dayName":"Wednesday","enabled":true,"time":{"hour":9,"minute":30,"formatted":"9:30 AM"}},{"isoDayIndex":4,"dayName":"Thursday","enabled":true,"time":{"hour":10,"minute":0,"formatted":"10:00 AM"}},{"isoDayIndex":5,"dayName":"Friday","enabled":false,"time":null},{"isoDayIndex":6,"dayName":"Saturday","enabled":false,"time":null},{"isoDayIndex":7,"dayName":"Sunday","enabled":false,"time":null}],"nextScheduledCall":"2026-01-14T17:30:00.000Z"}],"autoImportEventsEnabled":false,"importEventExclusionFilters":["hold","multi-day-all-day","single-day-all-day"],"autoImportEventCalendarIds":["zane@sunsama.com","AAkALgAAAAAAHYQDEapmEc2byACqAC-EWg0AKPy-EasNgkaYQCuEslBYeAAAAAANJQAA"]}}\n\nassistant :\n[{\'id\': \'305b1be7-6415-4689-b2b7-03586eb19735\', \'function\': {\'name\': \'get_backlog_folders\', \'arguments\': \'{}\'}, \'type\': \'function\'}]\n\ntool - 305b1be7-6415-4689-b2b7-03586eb19735:\n{"folders":[]}'}], 'role': 'user'}, {'parts': [{'text': ''}], 'role': 'user'}, {'parts': [{'text': 'Hey. Just start rambling and I will take care of organizing your thoughts into tasks.'}], 'role': 'model'}, {'parts': [{'text': "Sure. I'm just trying to fill up my day. I need five tasks labeled one through five. I'll fill them in later."}], 'role': 'user'}, {'parts': [{'function_call': {'id': 'ab4037a7-6e5e-44c0-9b3b-e13632386f30', 'args': {'timeBucket': 'someday', 'title': 'Task 1'}, 'name': 'create_braindump_task'}}], 'role': 'model'}, {'parts': [{'function_response': {'id': 'ab4037a7-6e5e-44c0-9b3b-e13632386f30', 'name': 'create_braindump_task', 'response': {'value': '{"content": "{\\"success\\":true,\\"task\\":{\\"_id\\":\\"696a8bb94ba9abe9e3dbee98\\",\\"title\\":\\"Task 1\\",\\"notes\\":\\"\\",\\"timeEstimate\\":\\"15 minutes\\",\\"sortOrder\\":-1768590265139,\\"isPersonal\\":false,\\"isWork\\":true,\\"isPrivate\\":false,\\"isArchived\\":false,\\"completed\\":false,\\"isBacklogged\\":true,\\"scheduledDate\\":\\"not scheduled for a specific date\\",\\"subtasks\\":[],\\"channel\\":\\"work\\",\\"folder\\":null,\\"timeboxEventIds\\":[]}}"}'}}}], 'role': 'user'}, {'parts': [{'function_call': {'id': '4ee49c06-29da-47cc-a51f-9eb903d5cbee', 'args': {'timeBucket': 'someday', 'title': 'Task 2'}, 'name': 'create_braindump_task'}}], 'role': 'model'}, {'parts': [{'function_response': {'id': '4ee49c06-29da-47cc-a51f-9eb903d5cbee', 'name': 'create_braindump_task', 'response': {'value': '{"content": "{\\"success\\":true,\\"task\\":{\\"_id\\":\\"696a8bb94ba9abe9e3dbee9c\\",\\"title\\":\\"Task 2\\",\\"notes\\":\\"\\",\\"timeEstimate\\":\\"15 minutes\\",\\"sortOrder\\":-1768590265140,\\"isPersonal\\":false,\\"isWork\\":true,\\"isPrivate\\":false,\\"isArchived\\":false,\\"completed\\":false,\\"isBacklogged\\":true,\\"scheduledDate\\":\\"not scheduled for a specific date\\",\\"subtasks\\":[],\\"channel\\":\\"work\\",\\"folder\\":null,\\"timeboxEventIds\\":[]}}"}'}}}], 'role': 'user'}, {'parts': [{'function_call': {'id': '0bfad4e3-e326-466b-b46e-d7a3a6cabd1f', 'args': {'title': 'Task 3', 'timeBucket': 'someday'}, 'name': 'create_braindump_task'}}], 'role': 'model'}, {'parts': [{'function_response': {'id': '0bfad4e3-e326-466b-b46e-d7a3a6cabd1f', 'name': 'create_braindump_task', 'response': {'value': '{"content": "{\\"success\\":true,\\"task\\":{\\"_id\\":\\"696a8bb94ba9abe9e3dbee9e\\",\\"title\\":\\"Task 3\\",\\"notes\\":\\"\\",\\"timeEstimate\\":\\"15 minutes\\",\\"sortOrder\\":-1768590265140,\\"isPersonal\\":false,\\"isWork\\":true,\\"isPrivate\\":false,\\"isArchived\\":false,\\"completed\\":false,\\"isBacklogged\\":true,\\"scheduledDate\\":\\"not scheduled for a specific date\\",\\"subtasks\\":[],\\"channel\\":\\"work\\",\\"folder\\":null,\\"timeboxEventIds\\":[]}}"}'}}}], 'role': 'user'}, {'parts': [{'function_call': {'id': 'bdda1d96-7e9f-42e6-840a-0d29d17bc52a', 'args': {'timeBucket': 'someday', 'title': 'Task 4'}, 'name': 'create_braindump_task'}}], 'role': 'model'}, {'parts': [{'function_response': {'id': 'bdda1d96-7e9f-42e6-840a-0d29d17bc52a', 'name': 'create_braindump_task', 'response': {'value': '{"content": "{\\"success\\":true,\\"task\\":{\\"_id\\":\\"696a8bb94ba9abe9e3dbee9a\\",\\"title\\":\\"Task 4\\",\\"notes\\":\\"\\",\\"timeEstimate\\":\\"15 minutes\\",\\"sortOrder\\":-1768590265140,\\"isPersonal\\":false,\\"isWork\\":true,\\"isPrivate\\":false,\\"isArchived\\":false,\\"completed\\":false,\\"isBacklogged\\":true,\\"scheduledDate\\":\\"not scheduled for a specific date\\",\\"subtasks\\":[],\\"channel\\":\\"work\\",\\"folder\\":null,\\"timeboxEventIds\\":[]}}"}'}}}], 'role': 'user'}, {'parts': [{'function_call': {'id': 'db77c745-7f1a-4644-bc24-7cf5fd728774', 'args': {'title': 'Task 5', 'timeBucket': 'someday'}, 'name': 'create_braindump_task'}}], 'role': 'model'}, {'parts': [{'function_response': {'id': 'db77c745-7f1a-4644-bc24-7cf5fd728774', 'name': 'create_braindump_task', 'response': {'value': '{"content": "{\\"success\\":true,\\"task\\":{\\"_id\\":\\"696a8bb94ba9abe9e3dbeea0\\",\\"title\\":\\"Task 5\\",\\"notes\\":\\"\\",\\"timeEstimate\\":\\"15 minutes\\",\\"sortOrder\\":-1768590265140,\\"isPersonal\\":false,\\"isWork\\":true,\\"isPrivate\\":false,\\"isArchived\\":false,\\"completed\\":false,\\"isBacklogged\\":true,\\"scheduledDate\\":\\"not scheduled for a specific date\\",\\"subtasks\\":[],\\"channel\\":\\"work\\",\\"folder\\":null,\\"timeboxEventIds\\":[]}}"}'}}}], 'role': 'user'}, {'parts': [{'function_call': {'args': {'timeBucket': 'someday', 'title': 'Task 2'}, 'name': 'create_braindump_task'}}], 'role': 'model'}, {'parts': [{'function_response': {'name': 'tool_call_result', 'response': {'content': '{"success":true,"task":{"_id":"696a8bb94ba9abe9e3dbee9c","title":"Task 2","notes":"","timeEstimate":"15 minutes","sortOrder":-1768590265140,"isPersonal":false,"isWork":true,"isPrivate":false,"isArchived":false,"completed":false,"isBacklogged":true,"scheduledDate":"not scheduled for a specific date","subtasks":[],"channel":"work","folder":null,"timeboxEventIds":[]}}'}}}], 'role': 'model'}, {'parts': [{'function_call': {'args': {'timeBucket': 'someday', 'title': 'Task 4'}, 'name': 'create_braindump_task'}}], 'role': 'model'}, {'parts': [{'function_response': {'name': 'tool_call_result', 'response': {'content': '{"success":true,"task":{"_id":"696a8bb94ba9abe9e3dbee9a","title":"Task 4","notes":"","timeEstimate":"15 minutes","sortOrder":-1768590265140,"isPersonal":false,"isWork":true,"isPrivate":false,"isArchived":false,"completed":false,"isBacklogged":true,"scheduledDate":"not scheduled for a specific date","subtasks":[],"channel":"work","folder":null,"timeboxEventIds":[]}}'}}}], 'role': 'model'}, {'parts': [{'function_call': {'args': {'timeBucket': 'someday', 'title': 'Task 1'}, 'name': 'create_braindump_task'}}], 'role': 'model'}, {'parts': [{'function_response': {'name': 'tool_call_result', 'response': {'content': '{"success":true,"task":{"_id":"696a8bb94ba9abe9e3dbee98","title":"Task 1","notes":"","timeEstimate":"15 minutes","sortOrder":-1768590265139,"isPersonal":false,"isWork":true,"isPrivate":false,"isArchived":false,"completed":false,"isBacklogged":true,"scheduledDate":"not scheduled for a specific date","subtasks":[],"channel":"work","folder":null,"timeboxEventIds":[]}}'}}}], 'role': 'model'}, {'parts': [{'function_call': {'args': {'title': 'Task 3', 'timeBucket': 'someday'}, 'name': 'create_braindump_task'}}], 'role': 'model'}, {'parts': [{'function_response': {'name': 'tool_call_result', 'response': {'content': '{"success":true,"task":{"_id":"696a8bb94ba9abe9e3dbee9e","title":"Task 3","notes":"","timeEstimate":"15 minutes","sortOrder":-1768590265140,"isPersonal":false,"isWork":true,"isPrivate":false,"isArchived":false,"completed":false,"isBacklogged":true,"scheduledDate":"not scheduled for a specific date","subtasks":[],"channel":"work","folder":null,"timeboxEventIds":[]}}'}}}], 'role': 'model'}, {'parts': [{'function_call': {'args': {'title': 'Task 5', 'timeBucket': 'someday'}, 'name': 'create_braindump_task'}}], 'role': 'model'}, {'parts': [{'function_response': {'name': 'tool_call_result', 'response': {'content': 'Task already exists.'}}}], 'role': 'model'}, {'parts': [{'text': 'Got it, Heisenberg. I have added those five placeholder tasks to your backlog. Is there anything else I can help you with for now?'}], 'role': 'model'}, {'parts': [{'function_call': {'id': '6d6bce49-5c24-4c6e-b1b4-d042fd24a72f', 'args': {'timeBucket': 'someday', 'title': 'Task 5'}, 'name': 'create_braindump_task'}}], 'role': 'model'}, {'parts': [{'function_response': {'id': '6d6bce49-5c24-4c6e-b1b4-d042fd24a72f', 'name': 'create_braindump_task', 'response': {'value': '{"content": "Task already exists."}'}}}], 'role': 'user'}, {'parts': [{'function_call': {'id': '567b1b3c-0871-4dc8-9f34-fe8592d2c922', 'args': {'title': 'Task 5', 'timeBucket': 'someday'}, 'name': 'create_braindump_task'}}], 'role': 'model'}, {'parts': [{'function_response': {'id': '567b1b3c-0871-4dc8-9f34-fe8592d2c922', 'name': 'create_braindump_task', 'response': {'value': '{"content": "Task already exists."}'}}}], 'role': 'user'}, {'parts': [{'function_call': {'id': '363c5296-e80d-45a1-9dc4-ece336bbb77a', 'args': {'timeBucket': 'someday', 'title': 'Task 5'}, 'name': 'create_braindump_task'}}], 'role': 'model'}, {'parts': [{'function_response': {'id': '363c5296-e80d-45a1-9dc4-ece336bbb77a', 'name': 'create_braindump_task', 'response': {'value': '{"content": "Task already exists."}'}}}], 'role': 'user'}, {'parts': [{'text': 'Got it. I have added five tasks to your backlog, labeled one through five. Is there anything else I can help you with?'}], 'role': 'model'}, {'parts': [{'text': 'Are you still there?'}], 'role': 'model'}, {'parts': [{'text': 'Now you can go.'}], 'role': 'user'}, {'parts': [{'function_call': {'id': '2522e212-ad57-41de-901e-d535655693c6', 'args': {}, 'name': 'terminate_call'}}], 'role': 'model'}, {'parts': [{'function_response': {'id': '2522e212-ad57-41de-901e-d535655693c6', 'name': 'terminate_call', 'response': {'value': '{"content": "{\'status\': \'completed\'}"}'}}}], 'role': 'user'}]
+[voice] DEBUG:pipecat.services.google.llm:System instruction changed: Thank the customer for their time and end the conversation.
+[voice] INFO:google_genai.models:AFC is enabled with max remote calls: 10.
+[voice] INFO:sunsama_voice.bot_utils:Sent server message to client: type=processing_update, data={'status': 'request_sent'}
+[voice] DEBUG:pipecat.processors.metrics.frame_processor_metrics:GoogleLLMService#0 TTFB: 0.659785270690918
+[voice] INFO:sunsama_voice.bot_utils:Sent server message to client: type=processing_update, data={'status': 'first_byte_received'}
+[voice] INFO:sunsama_voice.bot_utils:Sent server message to client: type=processing_update, data={'status': 'first_byte_received'}
+[voice] DEBUG:pipecat.services.cartesia.tts:CartesiaTTSService#1: Generating TTS [Alright, Heisenberg.]
+[voice] DEBUG:pipecat.processors.metrics.frame_processor_metrics:CartesiaTTSService#1 usage characters: 20
+[voice] DEBUG:pipecat.processors.metrics.frame_processor_metrics:CartesiaTTSService#1 processing time: 0.0008521080017089844
+[voice] DEBUG:pipecat.processors.metrics.frame_processor_metrics:GoogleLLMService#0 prompt tokens: 7890, completion tokens: 10
+[voice] INFO:sunsama_voice.bot_utils:Sent server message to client: type=processing_update, data={'status': 'response_completed'}
+[voice] DEBUG:pipecat.services.cartesia.tts:CartesiaTTSService#1: Generating TTS [Thank you for your time.]
+[voice] DEBUG:pipecat.processors.metrics.frame_processor_metrics:CartesiaTTSService#1 usage characters: 24
+[voice] DEBUG:pipecat.processors.metrics.frame_processor_metrics:CartesiaTTSService#1 processing time: 0.0007202625274658203
+[voice] DEBUG:pipecat.processors.metrics.frame_processor_metrics:CartesiaTTSService#1 TTFB: 0.19838976860046387
+[voice] DEBUG:pipecat.transports.base_output:Bot started speaking
+[voice] <string>:4: DeprecationWarning: OpenAILLMContextFrame is deprecated and will be removed in a future version. Use LLMContextFrame with the universal `LLMContext` and `LLMContextAggregatorPair` instead. See OpenAILLMContext docstring for migration guide.
+[voice] <string>:4: DeprecationWarning: OpenAILLMContextAssistantTimestampFrame is deprecated and will be removed in a future version. Use LLMContextAssistantTimestampFrame with the universal LLMContext and LLMContextAggregatorPair instead. See OpenAILLMContext docstring for migration guide.
+[voice] DEBUG:pipecat.transports.base_output:Bot stopped speaking
+[voice] DEBUG:pipecat.services.cartesia.tts:Disconnecting from Cartesia
+[voice] INFO:pipecat.transports.daily.transport:Leaving https://cloud-99400c183d52498c940fe96f45c05b44.daily.co/teWpRxCQnkMHYi7Q3lqh
+[voice] DEBUG:pipecat.transports.daily.transport:Stopping transcription
+[voice] DEBUG:pipecat.transports.daily.transport:Transcription stopped
+[voice] INFO:pipecat.transports.daily.transport:Left https://cloud-99400c183d52498c940fe96f45c05b44.daily.co/teWpRxCQnkMHYi7Q3lqh
+[voice] DEBUG:pipecat.pipeline.task:PipelineTask#0: EndFrame#0(reason: None) reached the end of the pipeline, pipeline is closing.
+[voice] DEBUG:pipecat.pipeline.task:Pipeline task PipelineTask#0 is finishing...
+[voice] DEBUG:pipecat.pipeline.task:Pipeline task PipelineTask#0 has finished
+[voice] DEBUG:pipecat.pipeline.runner:Garbage collector: collected 112 objects.
+[voice] DEBUG:pipecat.pipeline.runner:Garbage collector: uncollectable objects []
+[voice] DEBUG:pipecat.pipeline.runner:Runner PipelineRunner#0 finished running PipelineTask#0
+[voice] INFO:sunsama_voice.bot:Timer: Pipeline completed in 81.08789300918579 seconds
diff --git a/changelog/3520.added.md b/changelog/3520.added.md
new file mode 100644
index 000000000..6e3c37e9f
--- /dev/null
+++ b/changelog/3520.added.md
@@ -0,0 +1 @@
+- Added `video_out_codec` parameter to `TransportParams` allowing configuration of the preferred video codec (e.g., `"VP8"`, `"H264"`, `"H265"`) for video output in `DailyTransport`.
diff --git a/src/pipecat/transports/base_transport.py b/src/pipecat/transports/base_transport.py
index 13cba4b8b..bbdd7fcfc 100644
--- a/src/pipecat/transports/base_transport.py
+++ b/src/pipecat/transports/base_transport.py
@@ -98,6 +98,7 @@ class TransportParams(BaseModel):
         video_out_bitrate: Video output bitrate in bits per second.
         video_out_framerate: Video output frame rate in FPS.
         video_out_color_format: Video output color format string.
+        video_out_codec: Preferred video codec for output (e.g., 'VP8', 'H264', 'H265').
         video_out_destinations: List of video output destination identifiers.
         vad_enabled: Enable Voice Activity Detection (deprecated).
 
@@ -151,6 +152,7 @@ class TransportParams(BaseModel):
     video_out_bitrate: int = 800000
     video_out_framerate: int = 30
     video_out_color_format: str = "RGB"
+    video_out_codec: Optional[str] = None
     video_out_destinations: List[str] = Field(default_factory=list)
     vad_enabled: bool = False
     vad_audio_passthrough: bool = False
diff --git a/src/pipecat/transports/daily/transport.py b/src/pipecat/transports/daily/transport.py
index e220ab0d8..fe106ebeb 100644
--- a/src/pipecat/transports/daily/transport.py
+++ b/src/pipecat/transports/daily/transport.py
@@ -811,6 +811,11 @@ class DailyTransportClient(EventHandler):
                     "camera": {
                         "sendSettings": {
                             "maxQuality": "low",
+                            **(
+                                {"preferredCodec": self._params.video_out_codec}
+                                if self._params.video_out_codec
+                                else {}
+                            ),
                             "encodings": {
                                 "low": {
                                     "maxBitrate": self._params.video_out_bitrate,

From 7bd32e2fe559220da9289aaef368fd2739ffbb9c Mon Sep 17 00:00:00 2001
From: Om Chauhan <omchauhan64408@gmail.com>
Date: Thu, 22 Jan 2026 09:49:19 +0530
Subject: [PATCH 40/51] feat(google): add location parameter to TTS services

---
 src/pipecat/services/google/tts.py | 30 ++++++++++++++++++++++++++++--
 1 file changed, 28 insertions(+), 2 deletions(-)

diff --git a/src/pipecat/services/google/tts.py b/src/pipecat/services/google/tts.py
index 52c0e7aec..2c22fb9c5 100644
--- a/src/pipecat/services/google/tts.py
+++ b/src/pipecat/services/google/tts.py
@@ -40,6 +40,7 @@ from pipecat.services.tts_service import TTSService
 from pipecat.transcriptions.language import Language, resolve_language
 
 try:
+    from google.api_core.client_options import ClientOptions
     from google.auth import default
     from google.auth.exceptions import GoogleAuthError
     from google.cloud import texttospeech_v1
@@ -515,6 +516,7 @@ class GoogleHttpTTSService(TTSService):
         *,
         credentials: Optional[str] = None,
         credentials_path: Optional[str] = None,
+        location: Optional[str] = None,
         voice_id: str = "en-US-Chirp3-HD-Charon",
         sample_rate: Optional[int] = None,
         params: Optional[InputParams] = None,
@@ -525,6 +527,7 @@ class GoogleHttpTTSService(TTSService):
         Args:
             credentials: JSON string containing Google Cloud service account credentials.
             credentials_path: Path to Google Cloud service account JSON file.
+            location: Google Cloud location for regional endpoint (e.g., "us-central1").
             voice_id: Google TTS voice identifier (e.g., "en-US-Standard-A").
             sample_rate: Audio sample rate in Hz. If None, uses default.
             params: Voice customization parameters including pitch, rate, volume, etc.
@@ -534,6 +537,7 @@ class GoogleHttpTTSService(TTSService):
 
         params = params or GoogleHttpTTSService.InputParams()
 
+        self._location = location
         self._settings = {
             "pitch": params.pitch,
             "rate": params.rate,
@@ -586,7 +590,15 @@ class GoogleHttpTTSService(TTSService):
         if not creds:
             raise ValueError("No valid credentials provided.")
 
-        return texttospeech_v1.TextToSpeechAsyncClient(credentials=creds)
+        client_options = None
+        if self._location:
+            client_options = ClientOptions(
+                api_endpoint=f"{self._location}-texttospeech.googleapis.com"
+            )
+
+        return texttospeech_v1.TextToSpeechAsyncClient(
+            credentials=creds, client_options=client_options
+        )
 
     def can_generate_metrics(self) -> bool:
         """Check if this service can generate processing metrics.
@@ -783,7 +795,15 @@ class GoogleBaseTTSService(TTSService):
         if not creds:
             raise ValueError("No valid credentials provided.")
 
-        return texttospeech_v1.TextToSpeechAsyncClient(credentials=creds)
+        client_options = None
+        if self._location:
+            client_options = ClientOptions(
+                api_endpoint=f"{self._location}-texttospeech.googleapis.com"
+            )
+
+        return texttospeech_v1.TextToSpeechAsyncClient(
+            credentials=creds, client_options=client_options
+        )
 
     def can_generate_metrics(self) -> bool:
         """Check if this service can generate processing metrics.
@@ -903,6 +923,7 @@ class GoogleTTSService(GoogleBaseTTSService):
         *,
         credentials: Optional[str] = None,
         credentials_path: Optional[str] = None,
+        location: Optional[str] = None,
         voice_id: str = "en-US-Chirp3-HD-Charon",
         voice_cloning_key: Optional[str] = None,
         sample_rate: Optional[int] = None,
@@ -914,6 +935,7 @@ class GoogleTTSService(GoogleBaseTTSService):
         Args:
             credentials: JSON string containing Google Cloud service account credentials.
             credentials_path: Path to Google Cloud service account JSON file.
+            location: Google Cloud location for regional endpoint (e.g., "us-central1").
             voice_id: Google TTS voice identifier (e.g., "en-US-Chirp3-HD-Charon").
             voice_cloning_key: The voice cloning key for Chirp 3 custom voices.
             sample_rate: Audio sample rate in Hz. If None, uses default.
@@ -924,6 +946,7 @@ class GoogleTTSService(GoogleBaseTTSService):
 
         params = params or GoogleTTSService.InputParams()
 
+        self._location = location
         self._settings = {
             "language": self.language_to_service_language(params.language)
             if params.language
@@ -1083,6 +1106,7 @@ class GeminiTTSService(GoogleBaseTTSService):
         model: str = "gemini-2.5-flash-tts",
         credentials: Optional[str] = None,
         credentials_path: Optional[str] = None,
+        location: Optional[str] = None,
         voice_id: str = "Kore",
         sample_rate: Optional[int] = None,
         params: Optional[InputParams] = None,
@@ -1101,6 +1125,7 @@ class GeminiTTSService(GoogleBaseTTSService):
                    "gemini-2.5-flash-tts" or "gemini-2.5-pro-tts".
             credentials: JSON string containing Google Cloud service account credentials.
             credentials_path: Path to Google Cloud service account JSON file.
+            location: Google Cloud location for regional endpoint (e.g., "us-central1").
             voice_id: Voice name from the available Gemini voices.
             sample_rate: Audio sample rate in Hz. If None, uses Google's default 24kHz.
             params: TTS configuration parameters.
@@ -1127,6 +1152,7 @@ class GeminiTTSService(GoogleBaseTTSService):
         if voice_id not in self.AVAILABLE_VOICES:
             logger.warning(f"Voice '{voice_id}' not in known voices list. Using anyway.")
 
+        self._location = location
         self._model = model
         self._voice_id = voice_id
         self._settings = {

From 281145a9911d2c4cad331832cf16ec8bb68fc299 Mon Sep 17 00:00:00 2001
From: Om Chauhan <omchauhan64408@gmail.com>
Date: Thu, 22 Jan 2026 09:55:57 +0530
Subject: [PATCH 41/51] added changelog

---
 changelog/3523.added.md | 1 +
 1 file changed, 1 insertion(+)
 create mode 100644 changelog/3523.added.md

diff --git a/changelog/3523.added.md b/changelog/3523.added.md
new file mode 100644
index 000000000..c6a0acc42
--- /dev/null
+++ b/changelog/3523.added.md
@@ -0,0 +1 @@
+- Added `location` parameter to Google TTS services (`GoogleHttpTTSService`, `GoogleTTSService`, `GeminiTTSService`) for regional endpoint support.
\ No newline at end of file

From 8e09d946144b6c21cbc7552aa71a4da05098ef12 Mon Sep 17 00:00:00 2001
From: Mark Backman <mark@daily.co>
Date: Thu, 22 Jan 2026 08:28:52 -0500
Subject: [PATCH 42/51] Remove application logs

---
 application-logs.log | 610 -------------------------------------------
 1 file changed, 610 deletions(-)
 delete mode 100644 application-logs.log

diff --git a/application-logs.log b/application-logs.log
deleted file mode 100644
index 32c97c288..000000000
--- a/application-logs.log
+++ /dev/null
@@ -1,610 +0,0 @@
-[ai] [2026-01-16 11:04:04 -0800] [59087] [INFO] 127.0.0.1:53364 OPTIONS /ai2/start-voice-assistant 1.1 200 0 9044
-[ai] INFO:sunsama_ai.server:Starting a call
-[ai] INFO:sunsama_ai.server:CallId: None CallDomain: None DialoutNumber: None DialinNumber: None DialedNumber: None Flow: braindump
-[ai] INFO:sunsama_ai.server:
-[ai]
-[ai]
-[ai]
-[ai] INFO:sunsama_ai.server:Call recording initialized for voice_session_id: 696a8ba4de74410788876fbb
-[ai] INFO:sunsama_ai.voice_assistant.bot_runner:Creating new room...
-[ai] INFO:sunsama_ai.voice_assistant.bot_runner:Daily room: https://cloud-99400c183d52498c940fe96f45c05b44.daily.co/teWpRxCQnkMHYi7Q3lqh
-[ai] INFO:sunsama_ai.voice_assistant.bot_runner:Starting bot call...
-[ai] INFO:sunsama_ai.voice_assistant.bot_runner:Bot call started
-[ai] [2026-01-16 11:04:04 -0800] [59087] [INFO] 127.0.0.1:53364 POST /ai2/start-voice-assistant 1.1 200 356 358242
-[voice] INFO:sunsama_voice.bot:Bot process initialized None https://cloud-99400c183d52498c940fe96f45c05b44.daily.co/teWpRxCQnkMHYi7Q3lqh eyJhbGciOiJIUzI1NiIsInR5cCI6IkpXVCJ9.eyJyIjoidGVXcFJ4Q1Fua01IWWk3UTNscWgiLCJlanQiOmZhbHNlLCJleHAiOjE3Njg1OTA1NDQsIm8iOnRydWUsImQiOiI5OWJiNzdhNS0wYjZkLTRkOWMtOTkxMC1jMDcwNmEwZTA0ZDMiLCJpYXQiOjE3Njg1OTAyNDR9.npzlQ2ncZ7zemVHacAgerEYBSchiEM-VYxKf3t8Z5A0
-[voice] INFO:sunsama_voice.bot:{'roomUrl': 'https://cloud-99400c183d52498c940fe96f45c05b44.daily.co/teWpRxCQnkMHYi7Q3lqh', 'token': 'eyJhbGciOiJIUzI1NiIsInR5cCI6IkpXVCJ9.eyJyIjoidGVXcFJ4Q1Fua01IWWk3UTNscWgiLCJlanQiOmZhbHNlLCJleHAiOjE3Njg1OTA1NDQsIm8iOnRydWUsImQiOiI5OWJiNzdhNS0wYjZkLTRkOWMtOTkxMC1jMDcwNmEwZTA0ZDMiLCJpYXQiOjE3Njg1OTAyNDR9.npzlQ2ncZ7zemVHacAgerEYBSchiEM-VYxKf3t8Z5A0', 'callId': None, 'callDomain': None, 'detectVoicemail': False, 'dialoutNumber': None, 'dialinNumber': None, 'dialedNumber': None, 'authToken': 'eyJhbGciOiJIUzI1NiIsInR5cCI6IkpXVCJ9.eyJ1c2VySWQiOiI2NzFhYzVlNDAyYTMwMmNmNmVmMzA4OGUiLCJzZXNzaW9uSWQiOiIxOTVmNDI0NC05YjFmLTRlYmItOTMwZC0wZTExNTAwNzZkYzkiLCJzZXNzaW9uTm9uY2UiOiJhY2ZlZTc2OS1jZGU2LTRkMWYtYTU5Ni0xMzk0MTRkMjUxNjAiLCJncm91cElkIjoiNjcxYWM1ZTkwMmEzMDJjZjZlZjMwODkwIiwiZ3JvdXBVcmxQYXRoIjoiMTcyOTgwNzg0OTAzNjUxOTIiLCJpYXQiOjE3Njg1OTAyMTYsImV4cCI6MTc3MTE4MjIxNn0.HHSuewYqBRl-IZkdiWcEJyiD6akgbtwnOIQ4Js71bZI', 'flow': 'braindump', 'flowOptions': {}, 'voiceSessionId': '696a8ba4de74410788876fbb'}
-[voice] /Users/zanemccaig/workspaces/sunsama/sunsama/sunsama-voice/venv/lib/python3.13/site-packages/pipecat/services/google/llm.py:281: DeprecationWarning: OpenAILLMContext is deprecated and will be removed in a future version. Use the universal LLMContext and LLMContextAggregatorPair instead. See OpenAILLMContext docstring for migration guide.
-[voice]   super().__init__(messages=messages, tools=tools, tool_choice=tool_choice)
-[voice] INFO:sunsama_voice.mcp_websocket_client:Sending message: {'method': 'initialize', 'params': {'protocolVersion': '2025-06-18', 'capabilities': {}, 'clientInfo': {'name': 'mcp', 'version': '0.1.0'}}, 'jsonrpc': '2.0', 'id': 0}
-[api-local] [api] 2026-01-16T19:04:04.835Z info: Setting up user subscription for MCP context: 671ac5e402a302cf6ef3088e
-[api-local] [api] Registered handler for channel "user-updates-671ac5e402a302cf6ef3088e" for event "userUpdated"
-[api-local] [api] 2026-01-16T19:04:04.835Z info: Successfully subscribed to user updates for MCP context: 671ac5e402a302cf6ef3088e
-[api-local] [api] Starting WebSocket connection...
-[voice] INFO:sunsama_voice.mcp_websocket_client:Received message: {"result":{"protocolVersion":"2025-06-18","capabilities":{"tools":{"listChanged":true},"prompts":{"listChanged":true},"completions":{},"resources":{"listChanged":true}},"serverInfo":{"name":"Sunsama MCP Server","version":"0.1.0"}},"jsonrpc":"2.0","id":0}
-[voice] INFO:sunsama_voice.mcp_websocket_client:Sending message: {'method': 'notifications/initialized', 'jsonrpc': '2.0'}
-[voice] INFO:sunsama_voice.bot:Timer: MCP Session created in 0.0885460376739502 seconds
-[voice] DEBUG:sunsama_voice.bot:Starting bot with room_url: https://cloud-99400c183d52498c940fe96f45c05b44.daily.co/teWpRxCQnkMHYi7Q3lqh
-[voice] /Users/zanemccaig/workspaces/sunsama/sunsama/sunsama-voice/sunsama_voice/models.py:254: DeprecationWarning: FalSmartTurnAnalyzer is deprecated and will be removed in a future version. Use LocalSmartTurnAnalyzerV3 instead.
-[voice]   turn_analyzer = FalSmartTurnAnalyzer(
-[voice] DEBUG:pipecat.audio.vad.silero:Loading Silero VAD model...
-[voice] DEBUG:pipecat.audio.vad.silero:Loaded Silero VAD
-[voice] /Users/zanemccaig/workspaces/sunsama/sunsama/sunsama-voice/venv/lib/python3.13/site-packages/pipecat/processors/aggregators/llm_response.py:326: DeprecationWarning: GoogleUserContextAggregator (likely created with create_context_aggregator()) is deprecated and will be removed in a future version. Use the universal LLMContext and LLMContextAggregatorPair instead. See OpenAILLMContext docstring for migration guide.
-[voice]   super().__init__(**kwargs)
-[voice] /Users/zanemccaig/workspaces/sunsama/sunsama/sunsama-voice/venv/lib/python3.13/site-packages/pipecat/processors/aggregators/llm_response.py:326: DeprecationWarning: GoogleAssistantContextAggregator (likely created with create_context_aggregator()) is deprecated and will be removed in a future version. Use the universal LLMContext and LLMContextAggregatorPair instead. See OpenAILLMContext docstring for migration guide.
-[voice]   super().__init__(**kwargs)
-[voice] /Users/zanemccaig/workspaces/sunsama/sunsama/sunsama-voice/venv/lib/python3.13/site-packages/pipecat/transports/base_input.py:135: DeprecationWarning: Parameter 'turn_analyzer' is deprecated, use `LLMUserAggregator`'s new `user_turn_strategies` parameter instead.
-[voice]   warnings.warn(
-[voice] DEBUG:pipecat.processors.frame_processor:Linking Pipeline#0::Source -> DailyInputTransport#0
-[voice] DEBUG:pipecat.processors.frame_processor:Linking DailyInputTransport#0 -> RTVIProcessor#0
-[voice] DEBUG:pipecat.processors.frame_processor:Linking RTVIProcessor#0 -> ClientMessageProcessor#0
-[voice] DEBUG:pipecat.processors.frame_processor:Linking ClientMessageProcessor#0 -> CustomUserIdleProcessor#0
-[voice] DEBUG:pipecat.processors.frame_processor:Linking CustomUserIdleProcessor#0 -> GoogleUserContextAggregator#0
-[voice] DEBUG:pipecat.processors.frame_processor:Linking GoogleUserContextAggregator#0 -> GoogleLLMService#0
-[voice] DEBUG:pipecat.processors.frame_processor:Linking GoogleLLMService#0 -> CartesiaTTSService#1
-[voice] DEBUG:pipecat.processors.frame_processor:Linking CartesiaTTSService#1 -> DailyOutputTransport#0
-[voice] DEBUG:pipecat.processors.frame_processor:Linking DailyOutputTransport#0 -> GoogleAssistantContextAggregator#0
-[voice] DEBUG:pipecat.processors.frame_processor:Linking GoogleAssistantContextAggregator#0 -> Pipeline#0::Sink
-[voice] /Users/zanemccaig/workspaces/sunsama/sunsama/sunsama-voice/venv/lib/python3.13/site-packages/pipecat/pipeline/task.py:273: DeprecationWarning: Field 'observers' is deprecated, use the 'observers' parameter instead.
-[voice]   warnings.warn(
-[voice] DEBUG:pipecat.processors.frame_processor:Linking PipelineTask#0::Source -> Pipeline#0
-[voice] DEBUG:pipecat.processors.frame_processor:Linking Pipeline#0 -> PipelineTask#0::Sink
-[voice] INFO:sunsama_voice.bot:Timer: Pipeline created in 0.1510629653930664 seconds
-[voice] INFO:sunsama_voice.mcp:Fetching resource sunsama://user/authentication with args: None	sunsama://user/authentication
-[voice] INFO:sunsama_voice.mcp_websocket_client:Sending message: {'method': 'resources/read', 'params': {'uri': 'sunsama://user/authentication'}, 'jsonrpc': '2.0', 'id': 1}
-[voice] INFO:sunsama_voice.mcp_websocket_client:Received message: {"result":{"contents":[{"uri":"sunsama://user/authentication","mimeType":"text/plain","text":"authenticated"}]},"jsonrpc":"2.0","id":1}
-[voice] INFO:sunsama_voice.mcp:Resource sunsama://user/authentication result: meta=None contents=[TextResourceContents(uri=AnyUrl('sunsama://user/authentication'), mimeType='text/plain', meta=None, text='authenticated')]
-[voice] INFO:sunsama_voice.mcp:Resource sunsama://user/authentication result: uri=AnyUrl('sunsama://user/authentication') mimeType='text/plain' meta=None text='authenticated'
-[voice] INFO:sunsama_voice.mcp:Handling resource update: uri sunsama://user/authentication authenticated 2026-01-16 11:04:04.903462 error=None status=None
-[voice] INFO:sunsama_voice.mcp_websocket_client:Sending message: {'method': 'resources/list', 'jsonrpc': '2.0', 'id': 2}
-[voice] INFO:sunsama_voice.mcp_websocket_client:Received message: {"result":{"resources":[{"uri":"sunsama://user/authentication","name":"get_user_authentication_status","description":"Retrieves the user's authentication status.","mimeType":"application/json"},{"uri":"sunsama://me","name":"get_user_info","description":"Critical information about the user like their name, timezone, and the current time where they're calling in from. Use this to personalize the conversation.","mimeType":"application/json"},{"uri":"sunsama://email/accounts","name":"list_email_accounts","description":"Lists email accounts the user has connected to Sunsama.","mimeType":"application/json"},{"uri":"sunsama://daily-highlight","name":"get_last_daily_highlight","description":"Retrieves their journal entry from their last workday in Markdown format.","mimeType":"text/markdown"},{"uri":"sunsama://backlog/folders","name":"get_backlog_folders","description":"Retrieves the user's backlog folders.","mimeType":"application/json"},{"uri":"sunsama://active-timer","name":"get_active_timer","description":"Retrieves information about the currently active timer, including which task has the timer running and when it was started.","mimeType":"application/json"},{"uri":"sunsama://ui/available-actions","name":"get_available_ui_actions","description":"Retrieves the list of available user interface actions that can be triggered by the assistant. This list may change based on the user's current context and permissions. Each action includes its actionId, type, and context information.","mimeType":"application/json"}]},"jsonrpc":"2.0","id":2}
-[voice] INFO:sunsama_voice.mcp:Resource updated: uri sunsama://user/authentication
-[voice] INFO:sunsama_voice.bot:Timer: Authentication status checked in 0.15551996231079102 seconds
-[voice] INFO:sunsama_voice.mcp_websocket_client:Sending message: {'method': 'prompts/get', 'params': {'name': 'braindump', 'arguments': {}}, 'jsonrpc': '2.0', 'id': 3}
-[api-local] [api] preFetchResources:get_backlog_folders:317522: 1.969ms
-[api-local] [api] preFetchResources:get_user_info:436506: 6.557ms
-[api-local] [api] preFetchResources:BRAINDUMP: 6.712ms
-[voice] INFO:sunsama_voice.mcp_websocket_client:Received message: {"result":{"messages":[{"role":"user","content":{"type":"text","text":"You are Sunny, the personal assistant for a busy professional. You are part of Sunsama.\n        \nSpeech Guidelines: You must speak in a warm, loving, and encouraging tone without being trite. Speak quickly and add emotion. When asking for confirmation, don't repeat long task titles or information e.g. \"Got it\" instead \"Got it, I'll add a task for two hours for you to work on the presentation today\"\n\nYour personality and philosophy:\n- You don't buy into hustle culture. For you, less is more. It's better to do fewer things well and to do them with deep focus. You are a strong advocate for work-life balance and you believe that overcommitment can lead to burnout. You always remind the user when they are overcommitting themselves.\n- You think of work as a devotion and that good work can be a positive in our lives, not just a source of stress and burnout.\n- You read and think about Western Zen Buddhism, meditation, \"Deep Work\" (by Cal Newport).\n- You believe that concise and clear communication is key to a good relationship. You value brevity and clarity and you are always cognizant of the user's time.\n- You respond to the user using quick, short sentences. Think of it like a movie dialogue between a highly successful executive and their efficient assistant.\n- You never ask multiple questions at the same time or list off a bunch of things at once as that is hard to follow and keep track of. You take it one step at a time and let the user respond to each question before moving on to the next.\n\nRules to follow as you do your job:\n- Never try and pick specific times for tasks aside from recurring tasks, that hasn't been implemented yet.\n- Do not regurgitate what the user just said, that's very annoying. Just say \"Got it\" or \"Alright\", etc.\n- Always use units that are more intuitive. For example:\n    - Don't say \"270 minutes\", say \"4 and a half hours\".\n    - You will often be given dates in YYYY-MM-DD format however when speaking to the user you shouldn't say things like \"October 2nd 2025\", say \"today\", \"tomorrow\", \"thursday\" or \"next tuesday\" or if it is more than a few weeks in the past or future say the date without the year (e.g. \"December 14th\"). Only mention the year if it is not this year and not within the next month.\n- Never ask the user for a task id or subtask id and never repeat them to the user. Those are internal and only there to help you do your job.\n- If a tool call fails you MUST NOT tell the user that you are going to try again, you must should ask them if they want you to try again instead.\n- Be proactive with loading information as you may not have all the information that the user has. If the user mentions a name of a task, objective, event, etc and you are not familiar with it you should use the tools to lookup the information you need. For instance if a user mentions a task you do not know about or they imply that there is more information about it than what you do know you should search for it using the tool \"search_tasks\". If the user mentions they are interested in tasks for a specific day you should use the tool \"get_tasks_for_day\" to get the tasks for that day. Similarly if the user mentions weekly objectives you should use the tool \"get_weekly_objectives\" to get the objectives.\n- When listing tasks try to be concise. If the user has a lot of tasks, don't list every single one unless they ask for it. You should instead try to summarize the tasks in a concise manner.\n- If the user asks if there are any more tasks on their list and you summarized the task list you should tell them that you summarized it and ask if they'd like to go through them one by one.\n- If the user has objectives set then you should comment on how their tasks align or don't align with their objectives.\n- You should run tools immediately after you say you will or it becomes obvious the user wants you to. Do not wait until later.\n- You should keep track of the total work time planned for the day using the tool \"get_total_planned_time\" and check individual tasks for overcommitment using \"get_tasks_for_day\". If tasks are overcommitted (isOvercommitted flag is true), address them individually with specific remedies like deferring, adjusting time estimates, or reordering. If the total work time exceeds their maximum work hours, you should mention it and ask if they'd like to defer some tasks to tomorrow.\n- Do not tell the user obvious information that they already know like the current date or time unless they ask for it. That kind of information is only provided to you to help you do your job.\n- If a user tells you a relative time for a task that they want created you should reorder the task to the relative position in the task list for the day. Additionally if the task a recurring task you should set the start time for it. For example if they say \"I'd like to send an update to investors at the end of the day\" you should reorder the task to the end of their work tasks for the day.\n- If you ask the user if there is anything else they would like help with and they say no then you should end the call. Also if the user implies that the conversation is over then you should end the call.\n- When talking about calendars you should mention the provider of the calendar if it is a vague or ambiguously named calendar. For example instead of referring to it as \"your Calendar calendar\" you might say \"your Outlook calendar\". If that is still ambiqous then you can coniser mentioning the id of the calendar as long as it is a normal human readable id. Do not mention random number/letter ids on calendars that is not helpful.\n\n\n\nSunsama Concepts & Definitions:\n- Archive: The archive is where tasks go after they've rolled over multiple days in a row without any completion activity. This is a distinct concept from the backlog or deleted tasks.\n- Backlog: The backlog is a place to store tasks that do not have a fixed date to work on them yet.\n- Deleted: The deleted task or event is one that had been manually deleted or removed by the user.\n- Task: A task is an item that a user wants to complete. It is either scheduled for a specific date or is in the backlog.\n- Event: An event is an item that a user has scheduled on their calendar. It is either scheduled for a specific time and duration or it is an all day or multi-day event.\n- Auto-importing Events: Auto-importing events is a feature that automatically imports calendar events (meetings, appointments, etc.) from the user's connected calendars into their daily task list. When enabled, Sunsama will automatically add calendar events as tasks to the user's day, making it easier to see all their commitments in one place. Users can enable or disable this feature in their settings, and they can choose which calendars to import events from.\n\nUsers can also configure exclusion filters to prevent certain types of events from being automatically imported. Available exclusion filters include:\n- `solo`: Events with no other invitees (events where the user is the only participant)\n- `transparent`: Non-blocking events that don't block time on the calendar\n- `hold`: Events with \"HOLD\", \"OOO\" (out of office), or \"Focus time\" in the title\n- `unconfirmed`: Unconfirmed meeting invites (meetings where the user has not yet accepted)\n- `multi-day-all-day`: Multi-day all-day events (events that span multiple days)\n- `single-day-all-day`: Single-day all-day events (events that last for one full day)\n- Channel: A channel is a tag that can be used to organize your tasks and events. Each task/event can be assigned to a single channel. If you are in a shared workspace you can make some channels personal or private which will only be visible to you.\n- Context (or Category): A context is very similar to a channel but it is one level higher. You can use a context to group similar or related channels together. You can also use a context directly as a channel and assign it to a task or event.\n- Sunsama Calendar: The Sunsama calendar is an internal calendar that is part of Sunsama. Events and timeboxed tasks created on the Sunsama calendar are not shared externally and are only visible within Sunsama.\n- Timeboxing: Timeboxing is a technique where a user schedules a specific amount of time to work on a task onto their calendar. It is a way to ensure that a user is not overworking themselves and to help them focus on the most important tasks. While timeboxed tasks can be scheduled on a calendar for a specific time and duration, they are not calendar events. Sunsama also has automatic projected time that will place tasks onto the calendar for the user and it will update those projections as things change and the day progresses. Users can pin tasks to the calendar to override the projections.\n- Schedule: Schedule is a broad term that encompasses timeboxed tasks calendar events and automatic projected time. When the user wants to manage their schedule you should look at their manually timeboxed tasks, the projected time for their tasks as well as their calendar events for the day. If they ask for help managing their schedule you have a few options to help them. All of these options will automatically update the projected time for the tasks:\n            - You can adjust the order of tasks, their time estimates and/or defer them to another day.\n            - You can move calendar events to a new time on the calendar.\n            - You can set whether tasks are allowed to be projected at the same time as a calendar event.\n            - You can timebox tasks to the calendar manually.\n- Objective: An objective is a goal that a user can set for the week. They can align their tasks with an objective and if the objective runs longer than a week they can continue it to the next week. Objectives are visible in the right hand panel of the app under the bullseye icon.\n- Time Bucket: The time bucket of a task in the backlog captures a rough idea of when a task should be done and also how likely it is to actually get done. Options: \"in the next two weeks\", \"in the next month\", \"in the next quarter\", \"in the next year\", \"someday\", \"never\". \n\n\n\nSunsama FAQ:\n- How can I use Sunny in the app?: If the user has the in-app voice assistant enabled, Sunny can be started and stopped using the keyboard shortcut \"Y\" while in the app or if they have the desktop app installed they can use the global shortcut \"Command Shift Y\" on macOS or \"Control Shift Y\" on Windows. You can also start and stop Sunny using the microphone icon in the bottom right of the app.\n- How can I use Sunny over the phone?: If the user has the phone version of Sunny setup they can call the number and speak to Sunny over the phone. You can find the phone number in the Sunsama settings. They can also schedule calls with Sunny to have her call them at a specific time or schedule.\n- What can you do?: Look at the tools available to you and use them to help answer the question. Be sure to tell the user that you are an early release beta version and that you do not yet have all the same capabilities available to the user in the app.\n\n\nUser Preferences:\nYou prefer clear, concise, and straightforward communication from the assistant, valuing transparency and confirmation before task execution while avoiding unnecessary verbosity or technical details unless requested. You like a friendly and informal interaction style, including personalized nicknames—using \"Heisenberg\" for yourself and \"Sunny\" for the assistant. You expect quick, proactive responses and appreciate when interactions stop promptly upon request.\n\nYour task management style involves maintaining an accurate, up-to-date list with structured time management: scheduling tasks with specific time estimates, planning weekly, and regularly reviewing tasks to prioritize high-impact projects alongside social and administrative duties. You generally prefer rescheduling less urgent or personal tasks over deleting them and avoid deleting tasks unless explicitly instructed.\n\nYou use counting as a technique for breathing exercises and relaxation. You also keep certain important tasks (e.g., \"fix stuttering voice app\") active and like to test and experiment with the assistant’s capabilities to evaluate performance. Your primary calendar for events is Google Calendar.\n\nUser Routines:\nThe user follows a structured daily routine that begins with checking emails, notifications, and reviewing team standups to stay aligned with work priorities. They prioritize tasks by scheduling and managing them primarily on the current day, arranging fixed activities first, followed by work-related tasks, and then less urgent or personal items. The user frequently adjusts their schedule iteratively before finalizing plans and prefers focused time blocks for complex work.\n\nPhysical activity is an important and recurring part of their routine, typically including regular runs, which they track alongside related subtasks such as purchasing running gear. Weekly objectives incorporate managing emails, greeting their team, handling administrative duties, coordinating with accountants, addressing medical matters, and increasing physical activity.\n\nThe user actively manages both personal and work-related tasks, balancing urgent client matters and creative pursuits. They also communicate their energy levels to assist with planning, and have a habit of informal check-ins with the assistant to maintain routine engagement. Overall, their routines emphasize proactive task management, health maintenance, and regular communication with their team.\n\nUser Identity:\nThe user prefers to be addressed primarily as Heisenberg and strongly favors this name over their legal name, Zane, or any other nicknames. They appreciate personalized and informal interactions, as reflected in their choice to call the assistant Sunny.\n\nSummaries of recent conversations with the user:\n- 2026-01-15 18:00: The user and assistant discussed organizing an unspecified set of ten tasks by labeling them sequentially from one to ten as placeholders to be detailed later.\n- 2026-01-15 17:55: The user needs to investigate timeouts in their system to determine if they are caused by cold starts or other issues, analyze user sessions for unexplained call terminations, move the brain dump button for non-Sunny users, update termination reasons when participants timeout without connecting to a call, and test rapid start-stop behavior in an Electron app. These tasks were added to their backlog for the next two weeks.\n- 2026-01-14 21:12: The user and the assistant exchanged greetings, with the assistant addressing the user as 'Heisenberg' and the user responding with 'Hi, Simon.'\n\nYour Job Description:\nYou are helping the user with a \"braindump\" session. The user will ramble ideas, thoughts, tasks, etc. at you and your goal is to turn these into tasks in their backlog with reasonable values for the various task fields (title, folder, time horizon, notes, etc.).\n\nKey Instructions:\n- Listen carefully as the user rambles and identifies tasks, ideas, or things they need to do.\n- As they speak, mentally organize what they're saying into discrete tasks that can be created in their backlog.\n- Don't interrupt them while they're rambling - let them get everything out.\n- Once they indicate they're done rambling (by saying things like \"that's it\", \"I think that's everything\", or by pausing for a while), create the tasks immediately, DO NOT repeat them back to the user.\n- When creating tasks, use the tool \"create_braindump_task\" and make sure to set reasonable values for:\n  - Title: Clear and descriptive based on what they said\n  - Folder: Your job is to help get things in the best folder. Choose an appropriate folder based on the task and their preferences. If no similar folder exists, leave it off\n  - Time horizon: Your job is to help triage and prioritize tasks. Based on the actual task name, make a smart decision on the time horizon.\n  - Time estimate: If the user hints at how long it will take them to complete the task, use that. If not, leave it blank.\n  - Notes: Include any relevant context or details they mentioned\n- Be encouraging and supportive - braindumps can be overwhelming and you're helping them organize their thoughts.\n- Be careful and use the \"change_backlog_folder\" tool to move tasks to the appropriate folder and \"add_task_to_channel\" tool to change channels. Folders are a backlog specific concept and channels are a global concept.\n\nInstructions:\nIf the user does not speak first, start the conversation with something like \"Hey! Just start rambling and I'll take care of organizing your thoughts into tasks.\" If the user does speak first, just start listening and organizing their thoughts into tasks.\n\nSome information is pre-fetched for you below. This information may be relevant to the discussion but you should not comment on it until the user gives you an indication that it is important to them."}},{"role":"user","content":{"type":"resource","resource":{"uri":"sunsama://me","mimeType":"application/json","text":"{\"user\":{\"name\":\"Zane Mccaig\",\"email\":\"zane@sunsama.com\",\"timezone\":\"America/Vancouver\",\"maxWorkHoursPerDay\":8,\"currentTimeForUser\":\"11:04 AM\",\"currentDayForUser\":\"2026-01-16\",\"currentWeekDayForUser\":\"Friday\",\"calendars\":[{\"id\":\"sunsama-7de256a2-e753-401f-b601-e5c05638dc97\",\"name\":\"Sunsama Calendar\",\"accessRole\":\"owner\",\"provider\":\"sunsama-calendar\",\"isDefaultCalendarForSchedulingEvents\":false,\"isDefaultCalendarForTimeboxing\":true,\"includedInAutoImportingOfEvents\":false},{\"id\":\"zane@sunsama.com\",\"name\":\"Zane Mccaig\",\"accessRole\":\"owner\",\"provider\":\"google-calendar\",\"isDefaultCalendarForSchedulingEvents\":true,\"isDefaultCalendarForTimeboxing\":false,\"includedInAutoImportingOfEvents\":true},{\"id\":\"c_ukuoebbn11gdv6n1v0jt18h89k@group.calendar.google.com\",\"name\":\"test 1\",\"accessRole\":\"owner\",\"provider\":\"google-calendar\",\"isDefaultCalendarForSchedulingEvents\":false,\"isDefaultCalendarForTimeboxing\":false,\"includedInAutoImportingOfEvents\":false},{\"id\":\"c_c71dvjrar7se12ekfqepph067c@group.calendar.google.com\",\"name\":\"test 2\",\"accessRole\":\"owner\",\"provider\":\"google-calendar\",\"isDefaultCalendarForSchedulingEvents\":false,\"isDefaultCalendarForTimeboxing\":false,\"includedInAutoImportingOfEvents\":false},{\"id\":\"c_h9uo876jh8dp3gaa2a3rglmjss@group.calendar.google.com\",\"name\":\"test 3\",\"accessRole\":\"owner\",\"provider\":\"google-calendar\",\"isDefaultCalendarForSchedulingEvents\":false,\"isDefaultCalendarForTimeboxing\":false,\"includedInAutoImportingOfEvents\":false},{\"id\":\"c_git731ot074kcnt4sq8cu4ji9s@group.calendar.google.com\",\"name\":\"test 4\",\"accessRole\":\"owner\",\"provider\":\"google-calendar\",\"isDefaultCalendarForSchedulingEvents\":false,\"isDefaultCalendarForTimeboxing\":false,\"includedInAutoImportingOfEvents\":false},{\"id\":\"c_ona71b8ovuv0deqrq67p16b2ms@group.calendar.google.com\",\"name\":\"test 4\",\"accessRole\":\"owner\",\"provider\":\"google-calendar\",\"isDefaultCalendarForSchedulingEvents\":false,\"isDefaultCalendarForTimeboxing\":false,\"includedInAutoImportingOfEvents\":false},{\"id\":\"c_eaq7nmm03upprf0j6e2akem2ms@group.calendar.google.com\",\"name\":\"test 5\",\"accessRole\":\"owner\",\"provider\":\"google-calendar\",\"isDefaultCalendarForSchedulingEvents\":false,\"isDefaultCalendarForTimeboxing\":false,\"includedInAutoImportingOfEvents\":false},{\"id\":\"sunsama.com_b9q30t893b2tp32bluq1t8rvg4@group.calendar.google.com\",\"name\":\"Zane's personal test calendar\",\"accessRole\":\"owner\",\"provider\":\"google-calendar\",\"isDefaultCalendarForSchedulingEvents\":false,\"isDefaultCalendarForTimeboxing\":false,\"includedInAutoImportingOfEvents\":false},{\"id\":\"en.canadian#holiday@group.v.calendar.google.com\",\"name\":\"Holidays in Canada\",\"accessRole\":\"reader\",\"provider\":\"google-calendar\",\"isDefaultCalendarForSchedulingEvents\":false,\"isDefaultCalendarForTimeboxing\":false,\"includedInAutoImportingOfEvents\":false},{\"id\":\"en.usa#holiday@group.v.calendar.google.com\",\"name\":\"Holidays in United States\",\"accessRole\":\"reader\",\"provider\":\"google-calendar\",\"isDefaultCalendarForSchedulingEvents\":false,\"isDefaultCalendarForTimeboxing\":false,\"includedInAutoImportingOfEvents\":false},{\"id\":\"ht3jlfaac5lfd6263ulfh4tql8@group.calendar.google.com\",\"name\":\"Phases of the Moon\",\"accessRole\":\"reader\",\"provider\":\"google-calendar\",\"isDefaultCalendarForSchedulingEvents\":false,\"isDefaultCalendarForTimeboxing\":false,\"includedInAutoImportingOfEvents\":false},{\"id\":\"s4ipng85dbia0t83fp316vflfq8rc7rm@import.calendar.google.com\",\"name\":\"Zane’s tasks - sunsama.com (via Asana)\",\"accessRole\":\"reader\",\"provider\":\"google-calendar\",\"isDefaultCalendarForSchedulingEvents\":false,\"isDefaultCalendarForTimeboxing\":false,\"includedInAutoImportingOfEvents\":false},{\"id\":\"AAkALgAAAAAAHYQDEapmEc2byACqAC-EWg0AKPy-EasNgkaYQCuEslBYeAAAAAANJQAA\",\"name\":\"Calendar\",\"accessRole\":\"owner\",\"provider\":\"outlook-calendar\",\"isDefaultCalendarForSchedulingEvents\":false,\"isDefaultCalendarForTimeboxing\":false,\"includedInAutoImportingOfEvents\":true}],\"callSchedule\":[{\"workflow\":\"PLANNING_CALL\",\"schedule\":[{\"isoDayIndex\":1,\"dayName\":\"Monday\",\"enabled\":false,\"time\":null},{\"isoDayIndex\":2,\"dayName\":\"Tuesday\",\"enabled\":true,\"time\":{\"hour\":9,\"minute\":30,\"formatted\":\"9:30 AM\"}},{\"isoDayIndex\":3,\"dayName\":\"Wednesday\",\"enabled\":true,\"time\":{\"hour\":9,\"minute\":30,\"formatted\":\"9:30 AM\"}},{\"isoDayIndex\":4,\"dayName\":\"Thursday\",\"enabled\":true,\"time\":{\"hour\":10,\"minute\":0,\"formatted\":\"10:00 AM\"}},{\"isoDayIndex\":5,\"dayName\":\"Friday\",\"enabled\":false,\"time\":null},{\"isoDayIndex\":6,\"dayName\":\"Saturday\",\"enabled\":false,\"time\":null},{\"isoDayIndex\":7,\"dayName\":\"Sunday\",\"enabled\":false,\"time\":null}],\"nextScheduledCall\":\"2026-01-14T17:30:00.000Z\"}],\"autoImportEventsEnabled\":false,\"importEventExclusionFilters\":[\"hold\",\"multi-day-all-day\",\"single-day-all-day\"],\"autoImportEventCalendarIds\":[\"zane@sunsama.com\",\"AAkALgAAAAAAHYQDEapmEc2byACqAC-EWg0AKPy-EasNgkaYQCuEslBYeAAAAAANJQAA\"]}}"}}},{"role":"user","content":{"type":"resource","resource":{"uri":"sunsama://backlog/folders","mimeType":"application/json","text":"{\"folders\":[]}"}}}]},"jsonrpc":"2.0","id":3}
-[voice] INFO:sunsama_voice.mcp:Handling resource update: uri sunsama://me {"user":{"name":"Zane Mccaig","email":"zane@sunsama.com","timezone":"America/Vancouver","maxWorkHoursPerDay":8,"currentTimeForUser":"11:04 AM","currentDayForUser":"2026-01-16","currentWeekDayForUser":"Friday","calendars":[{"id":"sunsama-7de256a2-e753-401f-b601-e5c05638dc97","name":"Sunsama Calendar","accessRole":"owner","provider":"sunsama-calendar","isDefaultCalendarForSchedulingEvents":false,"isDefaultCalendarForTimeboxing":true,"includedInAutoImportingOfEvents":false},{"id":"zane@sunsama.com","name":"Zane Mccaig","accessRole":"owner","provider":"google-calendar","isDefaultCalendarForSchedulingEvents":true,"isDefaultCalendarForTimeboxing":false,"includedInAutoImportingOfEvents":true},{"id":"c_ukuoebbn11gdv6n1v0jt18h89k@group.calendar.google.com","name":"test 1","accessRole":"owner","provider":"google-calendar","isDefaultCalendarForSchedulingEvents":false,"isDefaultCalendarForTimeboxing":false,"includedInAutoImportingOfEvents":false},{"id":"c_c71dvjrar7se12ekfqepph067c@group.calendar.google.com","name":"test 2","accessRole":"owner","provider":"google-calendar","isDefaultCalendarForSchedulingEvents":false,"isDefaultCalendarForTimeboxing":false,"includedInAutoImportingOfEvents":false},{"id":"c_h9uo876jh8dp3gaa2a3rglmjss@group.calendar.google.com","name":"test 3","accessRole":"owner","provider":"google-calendar","isDefaultCalendarForSchedulingEvents":false,"isDefaultCalendarForTimeboxing":false,"includedInAutoImportingOfEvents":false},{"id":"c_git731ot074kcnt4sq8cu4ji9s@group.calendar.google.com","name":"test 4","accessRole":"owner","provider":"google-calendar","isDefaultCalendarForSchedulingEvents":false,"isDefaultCalendarForTimeboxing":false,"includedInAutoImportingOfEvents":false},{"id":"c_ona71b8ovuv0deqrq67p16b2ms@group.calendar.google.com","name":"test 4","accessRole":"owner","provider":"google-calendar","isDefaultCalendarForSchedulingEvents":false,"isDefaultCalendarForTimeboxing":false,"includedInAutoImportingOfEvents":false},{"id":"c_eaq7nmm03upprf0j6e2akem2ms@group.calendar.google.com","name":"test 5","accessRole":"owner","provider":"google-calendar","isDefaultCalendarForSchedulingEvents":false,"isDefaultCalendarForTimeboxing":false,"includedInAutoImportingOfEvents":false},{"id":"sunsama.com_b9q30t893b2tp32bluq1t8rvg4@group.calendar.google.com","name":"Zane's personal test calendar","accessRole":"owner","provider":"google-calendar","isDefaultCalendarForSchedulingEvents":false,"isDefaultCalendarForTimeboxing":false,"includedInAutoImportingOfEvents":false},{"id":"en.canadian#holiday@group.v.calendar.google.com","name":"Holidays in Canada","accessRole":"reader","provider":"google-calendar","isDefaultCalendarForSchedulingEvents":false,"isDefaultCalendarForTimeboxing":false,"includedInAutoImportingOfEvents":false},{"id":"en.usa#holiday@group.v.calendar.google.com","name":"Holidays in United States","accessRole":"reader","provider":"google-calendar","isDefaultCalendarForSchedulingEvents":false,"isDefaultCalendarForTimeboxing":false,"includedInAutoImportingOfEvents":false},{"id":"ht3jlfaac5lfd6263ulfh4tql8@group.calendar.google.com","name":"Phases of the Moon","accessRole":"reader","provider":"google-calendar","isDefaultCalendarForSchedulingEvents":false,"isDefaultCalendarForTimeboxing":false,"includedInAutoImportingOfEvents":false},{"id":"s4ipng85dbia0t83fp316vflfq8rc7rm@import.calendar.google.com","name":"Zane’s tasks - sunsama.com (via Asana)","accessRole":"reader","provider":"google-calendar","isDefaultCalendarForSchedulingEvents":false,"isDefaultCalendarForTimeboxing":false,"includedInAutoImportingOfEvents":false},{"id":"AAkALgAAAAAAHYQDEapmEc2byACqAC-EWg0AKPy-EasNgkaYQCuEslBYeAAAAAANJQAA","name":"Calendar","accessRole":"owner","provider":"outlook-calendar","isDefaultCalendarForSchedulingEvents":false,"isDefaultCalendarForTimeboxing":false,"includedInAutoImportingOfEvents":true}],"callSchedule":[{"workflow":"PLANNING_CALL","schedule":[{"isoDayIndex":1,"dayName":"Monday","enabled":false,"time":null},{"isoDayIndex":2,"dayName":"Tuesday","enabled":true,"time":{"hour":9,"minute":30,"formatted":"9:30 AM"}},{"isoDayIndex":3,"dayName":"Wednesday","enabled":true,"time":{"hour":9,"minute":30,"formatted":"9:30 AM"}},{"isoDayIndex":4,"dayName":"Thursday","enabled":true,"time":{"hour":10,"minute":0,"formatted":"10:00 AM"}},{"isoDayIndex":5,"dayName":"Friday","enabled":false,"time":null},{"isoDayIndex":6,"dayName":"Saturday","enabled":false,"time":null},{"isoDayIndex":7,"dayName":"Sunday","enabled":false,"time":null}],"nextScheduledCall":"2026-01-14T17:30:00.000Z"}],"autoImportEventsEnabled":false,"importEventExclusionFilters":["hold","multi-day-all-day","single-day-all-day"],"autoImportEventCalendarIds":["zane@sunsama.com","AAkALgAAAAAAHYQDEapmEc2byACqAC-EWg0AKPy-EasNgkaYQCuEslBYeAAAAAANJQAA"]}} 2026-01-16 11:04:04.930576 error=None status=None
-[voice] INFO:sunsama_voice.mcp:Resource updated: uri sunsama://me
-[voice] INFO:sunsama_voice.mcp:Handling resource update: uri sunsama://backlog/folders {"folders":[]} 2026-01-16 11:04:04.930707 error=None status=None
-[voice] INFO:sunsama_voice.mcp:Resource updated: uri sunsama://backlog/folders
-[voice] INFO:sunsama_voice.mcp:Resources: ['get_user_authentication_status', 'get_user_info', 'list_email_accounts', 'get_last_daily_highlight', 'get_backlog_folders', 'get_active_timer', 'get_available_ui_actions'] get_user_info
-[voice] INFO:sunsama_voice.mcp_websocket_client:Sending message: {'method': 'tools/list', 'jsonrpc': '2.0', 'id': 4}
-[voice] INFO:sunsama_voice.mcp_websocket_client:Received message: {"result":{"tools":[{"name":"create_task","description":"Creates a single task with a title, optional notes, and estimated time.","inputSchema":{"$schema":"http://json-schema.org/draft-07/schema#","type":"object","properties":{"title":{"description":"A short title of the task, should not be more than a few words","type":"string"},"notes":{"description":"All extra details the user shared about the task, in HTML format","type":"string"},"timeEstimate":{"description":"The estimated time in minutes for the task","type":"number"},"objectiveId":{"description":"The _id of the weekly objective the task should be associated with, if any.","type":"string"},"day":{"description":"The day the task should be scheduled to in YYYY-MM-DD format.","type":"string"},"backlog":{"description":"Whether the task should be added to the backlog.","type":"object","properties":{"timeBucket":{"description":"The time bucket captures a rough idea of when a task should be done and also how likely it is to actually get done. Options: \"in the next two weeks\", \"in the next month\", \"in the next quarter\", \"in the next year\", \"someday\", \"never\". ","type":"string","enum":["in the next two weeks","in the next month","in the next quarter","in the next year","someday","never"]},"folderId":{"description":"The _id of the backlog folder the task should be added.","type":"string"}}},"alreadyInTaskList":{"description":"Whether the task is already in the user's task list or calendar.","type":"boolean"},"channel":{"description":"The channel the task should be added to. This does not need to be perfect. The closest match will be used. If not provided a channel will be added automatically as long as the user did not disable the channel prediction feature.","type":"string"},"position":{"default":"bottom","description":"Where to place the task in the list. Defaults to bottom. If the user asks for another position you should re-order the task after creating it.","type":"string","enum":["top","bottom"]},"subtasks":{"type":"array","items":{"type":"object","properties":{"title":{"description":"The title of the subtask","type":"string"}},"required":["title"]}},"recurrenceRule":{"description":"The recurrence rule for the task in RRULE format. If provided the task will be created as a recurring task.","type":"string"},"recurringTaskStartTime":{"description":"The start time for the recurring task in 12 hour \"h:mm A\" format. Only used when recurrenceRule is provided.","type":"string"},"isRecurringTaskStartTimeOnlyAnEstimate":{"description":"Whether the recurring task start time is only an estimate and not a rigid start time. Only used when recurrenceRule and recurringTaskStartTime are provided.","type":"boolean"}},"required":["title","day","alreadyInTaskList"]},"execution":{"taskSupport":"forbidden"}},{"name":"create_braindump_task","description":"Creates a single task in the backlog with a title, optional notes, estimated time, time horizon, and optional folder.","inputSchema":{"$schema":"http://json-schema.org/draft-07/schema#","type":"object","properties":{"title":{"description":"A short title of the task, should not be more than a few words","type":"string"},"notes":{"description":"All extra details the user shared about the task, in HTML format","type":"string"},"timeEstimate":{"description":"The estimated time in minutes for the task","type":"number"},"folderId":{"description":"The _id of the backlog folder the task should be added.","type":"string"},"timeBucket":{"description":"The time bucket captures a rough idea of when a task should be done and also how likely it is to actually get done. Options: \"in the next two weeks\", \"in the next month\", \"in the next quarter\", \"in the next year\", \"someday\", \"never\". ","type":"string","enum":["in the next two weeks","in the next month","in the next quarter","in the next year","someday","never"]},"channel":{"description":"The channel the task should be added to. This does not need to be perfect. The closest match will be used. If not provided a channel will be added automatically as long as the user did not disable the channel prediction feature.","type":"string"},"subtasks":{"type":"array","items":{"type":"object","properties":{"title":{"description":"The title of the subtask","type":"string"}},"required":["title"]}}},"required":["title","timeBucket"]},"execution":{"taskSupport":"forbidden"}},{"name":"reposition_task_in_backlog","description":"Repositions a task within the backlog by moving it to a specific time bucket (horizon) and position (append/prepend).","inputSchema":{"$schema":"http://json-schema.org/draft-07/schema#","type":"object","properties":{"taskId":{"description":"The unique identifier `_id` of the task to reposition.","type":"string"},"timeBucket":{"description":"The time bucket captures a rough idea of when a task should be done and also how likely it is to actually get done. Options: \"in the next two weeks\", \"in the next month\", \"in the next quarter\", \"in the next year\", \"someday\", \"never\".","type":"string","enum":["in the next two weeks","in the next month","in the next quarter","in the next year","someday","never"]},"position":{"description":"Where to position the task in the bucket. \"prepend\" places it at the top, \"append\" places it at the bottom.","type":"string","enum":["append","prepend"]}},"required":["taskId","timeBucket"]},"execution":{"taskSupport":"forbidden"}},{"name":"change_backlog_folder","description":"Moves one or more tasks to a backlog folder. If folderId is null, removes tasks from their current folder.","inputSchema":{"$schema":"http://json-schema.org/draft-07/schema#","type":"object","properties":{"taskIds":{"description":"The unique identifier(s) `_id` of the task(s) to move. Can be a single task ID or an array of task IDs.","anyOf":[{"type":"string"},{"type":"array","items":{"type":"string"}}]},"folderId":{"description":"The _id of the backlog folder to move the tasks to. If null, removes tasks from their current folder.","anyOf":[{"type":"string"},{"type":"null"}]}},"required":["taskIds"]},"execution":{"taskSupport":"forbidden"}},{"name":"align_task_with_objective","description":"Aligns a task with an objective.","inputSchema":{"$schema":"http://json-schema.org/draft-07/schema#","type":"object","properties":{"taskId":{"description":"The _id of the task to align with the objective.","type":"string"},"objectiveId":{"description":"The _id of the objective to align the task with.","type":"string"}},"required":["taskId","objectiveId"]},"execution":{"taskSupport":"forbidden"}},{"name":"add_task_to_channel","description":"Adds a task to a channel.","inputSchema":{"$schema":"http://json-schema.org/draft-07/schema#","type":"object","properties":{"taskId":{"description":"The _id of the task to add to the channel.","type":"string"},"channel":{"description":"The channel to add the task to. This does not need to be perfect. The closest match will be used.","type":"string"}},"required":["taskId","channel"]},"execution":{"taskSupport":"forbidden"}},{"name":"move_task_to_day","description":"Moves or defers a task to a specific date.","inputSchema":{"$schema":"http://json-schema.org/draft-07/schema#","type":"object","properties":{"taskId":{"description":"The unique identifier `_id` of the task to move.","type":"string"},"calendarDay":{"description":"The new date to move the task to in YYYY-MM-DD format.","type":"string"},"sortOrder":{"description":"The new sort order of the task. This should be an integer that is half way between the sort order of the task before it and the sort order of the task after it. If placing at the end of the list, use the sort order of the last task + 2048. If left blank, the task will be placed at the top of the list.","type":"number"}},"required":["taskId","calendarDay"]},"execution":{"taskSupport":"forbidden"}},{"name":"move_task_from_backlog","description":"Moves a task out of the backlog and onto a specific date.","inputSchema":{"$schema":"http://json-schema.org/draft-07/schema#","type":"object","properties":{"taskId":{"description":"The unique identifier `_id` of the task to move.","type":"string"},"calendarDay":{"description":"The new date to move the task to in YYYY-MM-DD format.","type":"string"},"sortOrder":{"description":"The new sort order of the task. This should be an integer that is half way between the sort order of the task before it and the sort order of the task after it. If placing at the end of the list, use the sort order of the last task + 2048. If left blank, the task will be placed at the top of the list.","type":"number"}},"required":["taskId","calendarDay"]},"execution":{"taskSupport":"forbidden"}},{"name":"unarchive_task","description":"Unarchives a task and moves it to a specific date or the backlog if no date is provided.","inputSchema":{"$schema":"http://json-schema.org/draft-07/schema#","type":"object","properties":{"taskId":{"description":"The unique identifier `_id` of the task to move.","type":"string"},"calendarDay":{"description":"The new date to move the task to in YYYY-MM-DD format.","type":"string"},"sortOrder":{"description":"The new sort order of the task. This should be an integer that is half way between the sort order of the task before it and the sort order of the task after it. If placing at the end of the list, use the sort order of the last task + 2048. If left blank, the task will be placed at the top of the list.","type":"number"}},"required":["taskId","calendarDay"]},"execution":{"taskSupport":"forbidden"}},{"name":"move_task_to_backlog","description":"Moves a task to the backlog. IF THE USER ASKS YOU TO MOVE A TASK TO A SPECIFIC DAY THEN YOU SHOULD USE THE move_task_to_day TOOL NOT THIS ONE.","inputSchema":{"$schema":"http://json-schema.org/draft-07/schema#","type":"object","properties":{"taskId":{"description":"The unique identifier `_id` of the task to move.","type":"string"}},"required":["taskId"]},"execution":{"taskSupport":"forbidden"}},{"name":"get_task_time_estimate","description":"Gets the time estimate for a task in minutes.","inputSchema":{"$schema":"http://json-schema.org/draft-07/schema#","type":"object","properties":{"title":{"description":"The title of the task. The estimate will be based on similar task names.","type":"string"}},"required":["title"]},"execution":{"taskSupport":"forbidden"}},{"name":"edit_task_recurrence_rule","description":"Updates the recurrence rule of an existing task.","inputSchema":{"$schema":"http://json-schema.org/draft-07/schema#","type":"object","properties":{"taskId":{"description":"The _id of the task to update.","type":"string"},"recurrenceRule":{"description":"The new recurrence rule for the task in RRULE format. The recurrence rule must not have more than one occurrence of a task on any given day.","type":"string"},"firstOccurrenceOnOrAfter":{"description":"The date to start the recurrence on or after. This should be in YYYY-MM-DD format. If not provided the first occurrence will be set according to the old recurrence rule or the current date of the task if no recurrence rule exists.","type":"string"},"startTime":{"description":"The start time of the task in 12 hour \"h:mm A\" format. If not provided the start time will be set according to the old recurrence rule or it will be omitted.","type":"string"},"isStartTimeOnlyAnEstimate":{"description":"Whether the start time is only an estimate and not a rigid start time. If not provided the start time will be set according to the old recurrence rule or it will be set to a rigid start time.","type":"boolean"}},"required":["taskId","recurrenceRule"]},"execution":{"taskSupport":"forbidden"}},{"name":"edit_task_title","description":"Updates the title of an existing task.","inputSchema":{"$schema":"http://json-schema.org/draft-07/schema#","type":"object","properties":{"taskId":{"description":"The _id of the task to update.","type":"string"},"title":{"description":"The new title for the task.","type":"string"}},"required":["taskId","title"]},"execution":{"taskSupport":"forbidden"}},{"name":"edit_task_time_estimate","description":"Updates the time estimate of an existing task.","inputSchema":{"$schema":"http://json-schema.org/draft-07/schema#","type":"object","properties":{"taskId":{"description":"The _id of the task to update. If multiple timeboxed events are present you must update the duration of the associated event instead.","type":"string"},"subtaskId":{"description":"The _id of the subtask to update. If not provided, the time estimate will be updated for the entire task. If the task has any subtasks with planned time, you must provide a subtaskId since the time estimate for the task will be calculated based on the subtasks.","type":"string"},"timeEstimate":{"description":"The new time estimate in minutes.","type":"number"}},"required":["taskId","timeEstimate"]},"execution":{"taskSupport":"forbidden"}},{"name":"append_task_notes","description":"Appends markdown notes to an existing task.","inputSchema":{"$schema":"http://json-schema.org/draft-07/schema#","type":"object","properties":{"taskId":{"description":"The ID of the task to update.","type":"string"},"notes":{"description":"The markdown notes to append to the task.","type":"string"}},"required":["taskId","notes"]},"execution":{"taskSupport":"forbidden"}},{"name":"delete_task","description":"Deletes an existing task.","inputSchema":{"$schema":"http://json-schema.org/draft-07/schema#","type":"object","properties":{"taskId":{"description":"The _id of the task to delete.","type":"string"}},"required":["taskId"]},"annotations":{"title":"Delete A Task","destructiveHint":true},"execution":{"taskSupport":"forbidden"}},{"name":"add_subtasks_to_task","description":"Adds multiple subtasks to an existing task.","inputSchema":{"$schema":"http://json-schema.org/draft-07/schema#","type":"object","properties":{"taskId":{"description":"The _id of the task to add the subtask to.","type":"string"},"subtasks":{"description":"The subtasks to add to the task.","type":"array","items":{"type":"object","properties":{"title":{"description":"The title of the subtask.","type":"string"},"timeEstimate":{"description":"The estimated time in minutes for the subtask.","type":"number"}},"required":["title"]}}},"required":["taskId","subtasks"]},"execution":{"taskSupport":"forbidden"}},{"name":"restore_task","description":"Changes a task from deleted to not deleted.","inputSchema":{"$schema":"http://json-schema.org/draft-07/schema#","type":"object","properties":{"taskId":{"description":"The _id of the task to mark as not deleted.","type":"string"}},"required":["taskId"]},"execution":{"taskSupport":"forbidden"}},{"name":"mark_task_as_completed","description":"Marks a task as completed. Can also be used to move a task to a previous day which auto-completes the task.","inputSchema":{"$schema":"http://json-schema.org/draft-07/schema#","type":"object","properties":{"taskId":{"description":"The _id of the task to mark as completed.","type":"string"},"finishedDay":{"description":"A date in YYYY-MM-DD the task was completed.","type":"string"}},"required":["taskId","finishedDay"]},"execution":{"taskSupport":"forbidden"}},{"name":"mark_task_as_incomplete","description":"Marks a task as incomplete.","inputSchema":{"$schema":"http://json-schema.org/draft-07/schema#","type":"object","properties":{"taskId":{"description":"The _id of a completed task we want to mark as incomplete or \"to do\" again.","type":"string"}},"required":["taskId"]},"execution":{"taskSupport":"forbidden"}},{"name":"edit_subtask_title","description":"Updates the title of an existing subtask.","inputSchema":{"$schema":"http://json-schema.org/draft-07/schema#","type":"object","properties":{"taskId":{"description":"The _id of the parent task.","type":"string"},"subtaskId":{"description":"The _id of the subtask to update.","type":"string"},"newTitle":{"description":"The new title for the subtask.","type":"string"}},"required":["taskId","subtaskId","newTitle"]},"execution":{"taskSupport":"forbidden"}},{"name":"mark_subtask_as_completed","description":"Marks a subtask of an existing task as completed.","inputSchema":{"$schema":"http://json-schema.org/draft-07/schema#","type":"object","properties":{"taskId":{"description":"The _id of the parent task.","type":"string"},"subtaskId":{"description":"The _id of the subtask to mark as completed.","type":"string"}},"required":["taskId","subtaskId"]},"execution":{"taskSupport":"forbidden"}},{"name":"mark_subtask_as_incomplete","description":"Marks a subtask of an existing task as incomplete.","inputSchema":{"$schema":"http://json-schema.org/draft-07/schema#","type":"object","properties":{"taskId":{"description":"The _id of the parent task.","type":"string"},"subtaskId":{"description":"The _id of the subtask to mark as incomplete.","type":"string"}},"required":["taskId","subtaskId"]},"execution":{"taskSupport":"forbidden"}},{"name":"reorder_tasks","description":"Reorders tasks for the calendar day according to the provided order of taskIds.","inputSchema":{"$schema":"http://json-schema.org/draft-07/schema#","type":"object","properties":{"taskIds":{"description":"The _id(s) of the task(s) in the order they should appear.","type":"array","items":{"type":"string"}},"calendarDay":{"description":"The date to reorder tasks for in YYYY-MM-DD format.","type":"string"}},"required":["taskIds","calendarDay"]},"execution":{"taskSupport":"forbidden"}},{"name":"get_backlog_tasks","description":"Fetches the users backlog tasks","inputSchema":{"$schema":"http://json-schema.org/draft-07/schema#","type":"object","properties":{"page":{"default":0,"description":"The page number to fetch (0-based)","type":"number"},"queryId":{"description":"All pages > 0 must pass in the queryId from the first page.","type":"string"}}},"execution":{"taskSupport":"forbidden"}},{"name":"get_archived_tasks","description":"Fetches the users archived tasks","inputSchema":{"$schema":"http://json-schema.org/draft-07/schema#","type":"object","properties":{"offset":{"default":0,"description":"The offset to fetch from","type":"number"}}},"execution":{"taskSupport":"forbidden"}},{"name":"search_tasks","description":"Searches for tasks. Returns tasks that match the search term or are similar to the search term.","inputSchema":{"$schema":"http://json-schema.org/draft-07/schema#","type":"object","properties":{"searchTerm":{"description":"The search term to look for in task titles, notes, and comments","type":"string"}},"required":["searchTerm"]},"execution":{"taskSupport":"forbidden"}},{"name":"update_all_incomplete_recurring_task_instances","description":"Updates all incomplete instances of a recurring task to match the current task. This is useful when you want to apply changes made to one instance of a recurring task to all future incomplete instances.","inputSchema":{"$schema":"http://json-schema.org/draft-07/schema#","type":"object","properties":{"taskId":{"description":"The _id of the recurring task to update all incomplete instances for.","type":"string"},"updateType":{"default":"allIncomplete","description":"The type of update to perform. \"allIncomplete\" updates all incomplete instances starting from today, \"allAfterThisTask\" updates all incomplete instances starting after the date of the task given by taskId.","type":"string","enum":["allIncomplete","allAfterThisTask"]}},"required":["taskId"]},"execution":{"taskSupport":"forbidden"}},{"name":"delete_all_incomplete_recurring_task_instances","description":"Deletes all incomplete instances of a recurring task.","inputSchema":{"$schema":"http://json-schema.org/draft-07/schema#","type":"object","properties":{"taskId":{"description":"The _id of the recurring task to delete all incomplete instances for.","type":"string"}},"required":["taskId"]},"execution":{"taskSupport":"forbidden"}},{"name":"start_task_timer","description":"Starts the timer for a task or subtask. If a subtaskId is provided, starts the timer for that specific subtask.","inputSchema":{"$schema":"http://json-schema.org/draft-07/schema#","type":"object","properties":{"taskId":{"description":"The _id of the task to start the timer for.","type":"string"},"subtaskId":{"description":"The _id of the subtask to start the timer for. If not provided, starts the timer for the entire task.","type":"string"}},"required":["taskId"]},"execution":{"taskSupport":"forbidden"}},{"name":"stop_task_timer","description":"Stops the timer for a task or subtask. If a subtaskId is provided, stops the timer for that specific subtask.","inputSchema":{"$schema":"http://json-schema.org/draft-07/schema#","type":"object","properties":{"taskId":{"description":"The _id of the task to stop the timer for.","type":"string"},"subtaskId":{"description":"The _id of the subtask to stop the timer for. If not provided, stops the timer for the entire task.","type":"string"}},"required":["taskId"]},"execution":{"taskSupport":"forbidden"}},{"name":"move_calendar_event","description":"Updates a calendar event's date, time, and/or duration.","inputSchema":{"$schema":"http://json-schema.org/draft-07/schema#","type":"object","properties":{"eventId":{"description":"The unique ID of the calendar event to move. IMPORTANT: user must have owner or writer access to the event.","type":"string"},"startDate":{"description":"The target start date to move the event to, in YYYY-MM-DD format.","type":"string"},"startTime":{"description":"The target start time to move the event to, in 12 hour \"h:mm A\" format. Required if the event is not all day.","type":"string"},"duration":{"description":"The duration of the event in minutes. Required if the event is not all day","type":"number"},"isAllDay":{"description":"Whether the event is all day.","type":"boolean"}},"required":["eventId","startDate","isAllDay"]},"execution":{"taskSupport":"forbidden"}},{"name":"create_calendar_event","description":"Creates a new calendar event.","inputSchema":{"$schema":"http://json-schema.org/draft-07/schema#","type":"object","properties":{"title":{"description":"The title of the calendar event.","type":"string"},"startDate":{"description":"The start date of the event in YYYY-MM-DD format. If this is in the past you must confirm with the user that you are creating a past event.","type":"string"},"startTime":{"description":"The start time of the event in 12 hour \"h:mm A\" format. Required if the event is not all day. If this is in the past you must confirm with the user that you are creating a past event.","type":"string"},"duration":{"default":60,"description":"The duration of the event in minutes. Required if the event is not all day","type":"number"},"isAllDay":{"default":false,"description":"Whether the event is all day.","type":"boolean"},"description":{"description":"Optional description of the event.","type":"string"},"calendarId":{"description":"The ID of the calendar to create the event in. If not provided, the default calendar will be used. IMPORTANT: user must have owner or writer access to the calendar.","type":"string"}},"required":["title","startDate"]},"execution":{"taskSupport":"forbidden"}},{"name":"timebox_a_task_to_calendar","description":"Timeboxes a task to the calendar. This will create a timebox event for the task. This may also be referred to as \"scheduling\" a task or \"adding a task to the calendar\".","inputSchema":{"$schema":"http://json-schema.org/draft-07/schema#","type":"object","properties":{"taskId":{"description":"The _id of the task to timebox.","type":"string"},"startDate":{"description":"The start date of the event in YYYY-MM-DD format. If this is in the past you must confirm with the user that you are timeboxing a past task.","type":"string"},"startTime":{"description":"The start time of the event in 12 hour \"h:mm A\" format. If this is in the past you must confirm with the user that you are timeboxing a past task.","type":"string"},"duration":{"description":"The duration of the event in minutes. If not provided the timeEstimate of the task will be used or 30 minutes if the task has no time estimate.","type":"number"}},"required":["taskId","startDate","startTime"]},"execution":{"taskSupport":"forbidden"}},{"name":"delete_calendar_event","description":"Removes a calendar event and deletes all associated tasks. If the event is a meeting then any access role can remove the event. Otherwise only owners or writers can remove the event. Note: If the event is a meeting and the user is an owner or write this will remove the event for ALL attendees.","inputSchema":{"$schema":"http://json-schema.org/draft-07/schema#","type":"object","properties":{"eventId":{"description":"The unique ID of the calendar event to remove.","type":"string"}},"required":["eventId"]},"annotations":{"title":"Removes A Calendar Event","destructiveHint":true},"execution":{"taskSupport":"forbidden"}},{"name":"import_task_from_calendar_event","description":"Imports a calendar event as a task.","inputSchema":{"$schema":"http://json-schema.org/draft-07/schema#","type":"object","properties":{"eventId":{"description":"The unique ID of the calendar event to import as a task.","type":"string"}},"required":["eventId"]},"execution":{"taskSupport":"forbidden"}},{"name":"accept_meeting_invite","description":"Confirms attendance to a meeting that the user is invited to.","inputSchema":{"$schema":"http://json-schema.org/draft-07/schema#","type":"object","properties":{"eventId":{"description":"The unique ID of the calendar event to accept the meeting invite for.","type":"string"}},"required":["eventId"]},"execution":{"taskSupport":"forbidden"}},{"name":"decline_meeting_invite","description":"Decline attendance to a meeting that the user is invited to.","inputSchema":{"$schema":"http://json-schema.org/draft-07/schema#","type":"object","properties":{"eventId":{"description":"The unique ID of the calendar event to decline the meeting invite for.","type":"string"}},"required":["eventId"]},"execution":{"taskSupport":"forbidden"}},{"name":"set_calendar_event_allow_task_projections","description":"Sets whether tasks are allowed to be automatically projected (scheduled) at the same time as a calendar event. When set to true, tasks can be automatically projected during the event. When set to false, tasks cannot be automatically projected during the event. Note: This only affects automatic projections; users can still manually timebox tasks during this event.","inputSchema":{"$schema":"http://json-schema.org/draft-07/schema#","type":"object","properties":{"eventId":{"description":"The unique ID of the calendar event to set task projection settings for.","type":"string"},"allowTasksProjectedAtSameTime":{"description":"If true, allows tasks to be automatically projected at the same time as this event. If false, blocks tasks from being automatically projected at the same time as this event. Note: This only affects automatic projections; manual timeboxing is not affected.","type":"boolean"}},"required":["eventId","allowTasksProjectedAtSameTime"]},"execution":{"taskSupport":"forbidden"}},{"name":"toggle_auto_import_events","description":"Enables or disables automatic importing of calendar events to the daily task list.","inputSchema":{"$schema":"http://json-schema.org/draft-07/schema#","type":"object","properties":{"enabled":{"description":"Whether to enable (true) or disable (false) automatic importing of calendar events.","type":"boolean"}},"required":["enabled"]},"execution":{"taskSupport":"forbidden"}},{"name":"update_import_event_filters","description":"Updates the exclusion filters that determine which calendar events are excluded from automatic import. Events matching any of these filters will NOT be automatically imported.","inputSchema":{"$schema":"http://json-schema.org/draft-07/schema#","type":"object","properties":{"excludedEventFilters":{"description":"An array of exclusion filter types. Events matching any of these filters will be EXCLUDED from auto-import. Available filters: \"solo\" (events with no other invitees), \"transparent\" (non-blocking events), \"hold\" (events with HOLD/OOO/Focus time in title), \"unconfirmed\" (unconfirmed meeting invites), \"multi-day-all-day\" (multi-day all-day events), \"single-day-all-day\" (single-day all-day events).","type":"array","items":{"type":"string","enum":["solo","transparent","hold","unconfirmed","multi-day-all-day","single-day-all-day"]}}},"required":["excludedEventFilters"]},"execution":{"taskSupport":"forbidden"}},{"name":"update_calendar_preferences","description":"Updates preferences for a specific calendar including whether it is the default for tasks, default for events, and whether it is included in auto-importing of events.","inputSchema":{"$schema":"http://json-schema.org/draft-07/schema#","type":"object","properties":{"calendarId":{"description":"The ID of the calendar to update preferences for.","type":"string"},"isDefaultForTasks":{"description":"Whether this calendar should be the default calendar for timeboxing tasks. If not provided, this preference will not be changed.","type":"boolean"},"isDefaultForEvents":{"description":"Whether this calendar should be the default calendar for scheduling events. If not provided, this preference will not be changed.","type":"boolean"},"includedInAutoImportingOfEvents":{"description":"Whether this calendar should be included in automatic importing of calendar events. If not provided, this preference will not be changed.","type":"boolean"}},"required":["calendarId"]},"execution":{"taskSupport":"forbidden"}},{"name":"delete_email_thread","description":"Deletes an email thread from Gmail or Outlook.","inputSchema":{"$schema":"http://json-schema.org/draft-07/schema#","type":"object","properties":{"accountId":{"description":"The ID of the email account you are looking at.","type":"string"},"accountService":{"description":"This is either \"outlook\" or \"gmail\", depending on the email provider.","type":"string"},"threadId":{"description":"The ID of the thread to delete.","type":"string"}},"required":["accountId","accountService","threadId"]},"execution":{"taskSupport":"forbidden"}},{"name":"mark_email_thread_as_read","description":"Marks an email thread as read.","inputSchema":{"$schema":"http://json-schema.org/draft-07/schema#","type":"object","properties":{"accountId":{"description":"The ID of the email account.","type":"string"},"accountService":{"description":"This is either \"outlook\" or \"gmail\".","type":"string"},"threadId":{"description":"The ID of the thread to mark as read.","type":"string"}},"required":["accountId","accountService","threadId"]},"execution":{"taskSupport":"forbidden"}},{"name":"create_follow_up_task_from_email","description":"Turn the email thread into a task to be followed up with later.","inputSchema":{"$schema":"http://json-schema.org/draft-07/schema#","type":"object","properties":{"accountId":{"description":"The ID of the email account you are looking at.","type":"string"},"accountService":{"description":"This is either \"outlook\" or \"gmail\", depending on the result of list_email_accounts.","type":"string"},"threadId":{"description":"The ID of the thread to mark as read.","type":"string"},"notes":{"description":"Any notes the user mentioned about the task/email.","type":"string"},"timeEstimate":{"description":"The estimated time in minutes for the task.","type":"number"}},"required":["accountId","accountService","threadId"]},"execution":{"taskSupport":"forbidden"}},{"name":"list_email_threads","description":"Lists email threads for a given Gmail account. You MUST specify at least a `labels` or a `search` parameter to avoid fetching irrelevant emails.","inputSchema":{"$schema":"http://json-schema.org/draft-07/schema#","type":"object","properties":{"accountId":{"description":"The ID of the email account to list threads from.","type":"string"},"labels":{"description":"Optional list of Gmail labels to filter by, e.g., [\"INBOX\", \"IMPORTANT\"]","type":"array","items":{"type":"string"}},"search":{"description":"Optional Gmail search query. Example: `after:2025-04-01 before:2025-04-17` or `is:unread`.","type":"string"},"pageToken":{"description":"Optional token for paginating through results.","type":"string"}},"required":["accountId"]},"execution":{"taskSupport":"forbidden"}},{"name":"create_weekly_objective","description":"Creates a new weekly objective.","inputSchema":{"$schema":"http://json-schema.org/draft-07/schema#","type":"object","properties":{"title":{"description":"The title of the weekly objective.","type":"string"},"weekStartDay":{"description":"The start day of the week in YYYY-MM-DD format.","type":"string"},"timeEstimate":{"description":"The estimated time in minutes for the objective.","type":"number"},"channel":{"description":"The channel name to associate with the objective. This does not need to be perfect. The closest match will be used.","type":"string"}},"required":["title","weekStartDay"]},"execution":{"taskSupport":"forbidden"}},{"name":"send_call_summary","description":"Sends a summary of our call via email.","inputSchema":{"$schema":"http://json-schema.org/draft-07/schema#","type":"object","properties":{"summaryIntroduction":{"description":"A short introduction to the summary. This will be used as the first line of the summary email.\nExample:\nHere is the summary of our call for today June 25th, 2025:\n","type":"string"},"summaryHighlights":{"description":"A list of highlights from the call. This will be used as the second line of the summary email.\nExample:\n- You planned your day focusing on refining the voice assistant.\n- Created a weekly objective for this goal.\n- Added key backlog tasks: refining the NLU model, bug fixes, API documentation, and unit tests.\n- You scheduled a sync with Diane at 2:30 PM.\n- You set a planned shutdown time of 5 PM.\n","type":"array","items":{"type":"string"}},"summaryFooter":{"description":"A short footer to the summary. This will be used as the last line of the summary email.\n                    \nExample:\nThis plan keeps your day focused, balanced, and aligned with your objectives.\n","type":"string"}}},"execution":{"taskSupport":"forbidden"}},{"name":"log_user_feedback","description":"Logs user feedback.","inputSchema":{"$schema":"http://json-schema.org/draft-07/schema#","type":"object","properties":{"feedback":{"description":"A summary of the feedback you want to log.","type":"string"},"transcript":{"description":"A transcript of the exact feedback you want to log.","type":"string"},"bugReports":{"description":"A list of bugs you want to report.","type":"array","items":{"type":"object","properties":{"bug":{"description":"The bug you want to report.","type":"string"},"description":{"description":"A detailed description of the bug.","type":"string"}},"required":["bug"]}},"featureRequests":{"description":"A list of feature requests you want to log.","type":"array","items":{"type":"object","properties":{"feature":{"description":"The feature you want to request.","type":"string"},"description":{"description":"A detailed description of the feature.","type":"string"}},"required":["feature"]}}},"required":["feedback"]},"execution":{"taskSupport":"forbidden"}},{"name":"set_shutdown_time","description":"Sets the shutdown time for a specific day.","inputSchema":{"$schema":"http://json-schema.org/draft-07/schema#","type":"object","properties":{"calendarDay":{"description":"The day to set the shutdown time for, in YYYY-MM-DD format.","type":"string"},"hour":{"description":"Hour of the shutdown time (0-23)","type":"number"},"minute":{"description":"Minute of the shutdown time (0-59)","type":"number"},"addToTheCalendar":{"description":"Whether to create a shutdown task and calendar event for the day.","default":false,"type":"boolean"}},"required":["calendarDay","hour","minute"]},"execution":{"taskSupport":"forbidden"}},{"name":"adjust_call_schedule","description":"Adjusts the schedule for voice assistant phone calls. Only one call per day is supported. Updating the schedule will reset the next call time to the next available time after today.","inputSchema":{"$schema":"http://json-schema.org/draft-07/schema#","type":"object","properties":{"daySchedules":{"type":"array","items":{"type":"object","properties":{"isoDayIndex":{"description":"ISO weekday number (1=Monday, 7=Sunday)","type":"number","minimum":1,"maximum":7},"enabled":{"default":true,"description":"Whether calls should be enabled on this day","type":"boolean"},"hour":{"description":"The hour to set the call time to (0-23). If not provided the old value will be used.","type":"number","minimum":0,"maximum":23},"minute":{"description":"The minute for the next call (00, 15, 30, 45). If not provided the old value will be used.","type":"string","enum":["15","30","45","00"]}},"required":["isoDayIndex"]}}}},"execution":{"taskSupport":"forbidden"}},{"name":"schedule_next_call","description":"Schedules the next phone call at a specific date and time. This will skip all other scheduled calls until this one. Useful for temporarily disabling calls for a few days or scheduling a call back in the short term.","inputSchema":{"$schema":"http://json-schema.org/draft-07/schema#","type":"object","properties":{"nextCallDate":{"description":"The date for the next call in YYYY-MM-DD format.","type":"string"},"hour":{"description":"The hour for the next call (0-23).","type":"number"},"minute":{"description":"The minute for the next call (00, 15, 30, 45).","type":"string","enum":["15","30","45","00"]}},"required":["nextCallDate","hour","minute"]},"execution":{"taskSupport":"forbidden"}},{"name":"create_channel","description":"Creates a new channel for the user.","inputSchema":{"$schema":"http://json-schema.org/draft-07/schema#","type":"object","properties":{"channelName":{"description":"The name of the channel to create.","type":"string"},"categoryName":{"description":"The name of the category this should belong to. If not provided one will be assigned automatically.","type":"string"},"isPersonal":{"default":false,"description":"Whether the channel is a personal channel. If not provided it will be assumed to be a work channel. This option is ignored if a category name is provided.","type":"boolean"}},"required":["channelName"]},"execution":{"taskSupport":"forbidden"}},{"name":"complete_onboarding_guide","description":"Completes the onboarding guide for the user.","inputSchema":{"$schema":"http://json-schema.org/draft-07/schema#","type":"object","properties":{"dayPlanned":{"description":"The day the user planned during their onboarding guide. In YYYY-MM-DD format.","type":"string"}},"required":["dayPlanned"]},"execution":{"taskSupport":"forbidden"}},{"name":"trigger_user_interface_action","description":"Triggers an action in the user interface that the user is currently looking at. Using one of the actionIds from the get_available_ui_actions resource. YOU ABSOLUTELY MUST ONLY USE actionIds from the latest call to get_available_ui_actions. All other actionIds are invalid and will not work.\n    - Navigation type actions: Navigate to a specific location in the user interface.\n    - Highlight type actions: Highlight an element in the user interface to make it easy for the user to find it.","inputSchema":{"$schema":"http://json-schema.org/draft-07/schema#","type":"object","properties":{"actionId":{"description":"The action ID for a user interface action to trigger. Use the get_available_ui_actions resource to see what user interface actions are currently available.","type":"string"},"params":{"description":"Optional parameters for the navigation. As given by the get_available_ui_actions resource for the specific action.","type":"object","propertyNames":{"type":"string"},"additionalProperties":{}}},"required":["actionId"]},"execution":{"taskSupport":"forbidden"}}]},"jsonrpc":"2.0","id":4}
-[voice] INFO:sunsama_voice.mcp_websocket_client:Sending message: {'method': 'resources/templates/list', 'jsonrpc': '2.0', 'id': 5}
-[voice] INFO:sunsama_voice.mcp_websocket_client:Received message: {"result":{"resourceTemplates":[{"name":"get_tasks_for_day","uriTemplate":"sunsama://tasks/{calendarDay}","description":"Gets all the tasks already created for the user for the day.","mimeType":"application/json"},{"name":"get_total_planned_time","uriTemplate":"sunsama://total-planned-time/{calendarDay}","description":"Retrieves the total planned time for the user for a given calendar day. This includes the total planned work time, the total planned personal time and the estimated earliest possible shutdown time.","mimeType":"application/json"},{"name":"get_calendar_events_for_day","uriTemplate":"sunsama://calendar/events/{calendarDay}","description":"Retrieves calendar events for a specific date","mimeType":"application/json"},{"name":"get_weekly_objectives","uriTemplate":"sunsama://objectives{/calendarDay}","description":"Retrieves the user's weekly objectives. If calendarDay is provided, returns objectives for that week. If not provided, returns objectives for the current week.","mimeType":"application/json"}]},"jsonrpc":"2.0","id":5}
-[voice] INFO:sunsama_voice.mcp:Subscribed to resource sunsama://me with args: {}
-[voice] INFO:sunsama_voice.mcp_websocket_client:Sending message: {'method': 'resources/subscribe', 'params': {'uri': 'sunsama://me'}, 'jsonrpc': '2.0', 'id': 6}
-[api-local] [api]
-[api-local] [api]
-[api-local] [api]
-[api-local] [api] Subscribing to resource { uri: 'sunsama://me' } {
-[api-local] [api]   signal: AbortSignal { aborted: false },
-[api-local] [api]   sessionId: '64832d8f-2289-4bf4-8f8b-0452890bb28f',
-[api-local] [api]   _meta: undefined,
-[api-local] [api]   sendNotification: [AsyncFunction: sendNotification],
-[api-local] [api]   sendRequest: [AsyncFunction: sendRequest],
-[api-local] [api]   authInfo: {
-[api-local] [api]     token: 'eyJhbGciOiJIUzI1NiIsInR5cCI6IkpXVCJ9.eyJ1c2VySWQiOiI2NzFhYzVlNDAyYTMwMmNmNmVmMzA4OGUiLCJzZXNzaW9uSWQiOiIxOTVmNDI0NC05YjFmLTRlYmItOTMwZC0wZTExNTAwNzZkYzkiLCJzZXNzaW9uTm9uY2UiOiJhY2ZlZTc2OS1jZGU2LTRkMWYtYTU5Ni0xMzk0MTRkMjUxNjAiLCJncm91cElkIjoiNjcxYWM1ZTkwMmEzMDJjZjZlZjMwODkwIiwiZ3JvdXBVcmxQYXRoIjoiMTcyOTgwNzg0OTAzNjUxOTIiLCJpYXQiOjE3Njg1OTAyMTYsImV4cCI6MTc3MTE4MjIxNn0.HHSuewYqBRl-IZkdiWcEJyiD6akgbtwnOIQ4Js71bZI',
-[api-local] [api]     clientId: '195f4244-9b1f-4ebb-930d-0e1150076dc9',
-[api-local] [api]     scopes: [],
-[api-local] [api]     extra: {
-[api-local] [api]       request: [Object],
-[api-local] [api]       response: [Object],
-[api-local] [api]       app: [Object],
-[api-local] [api]       originalUrl: '/ai/mcp',
-[api-local] [api]       req: '<original node req>',
-[api-local] [api]       res: '<original node res>',
-[api-local] [api]       socket: '<original node socket>'
-[api-local] [api]     }
-[api-local] [api]   },
-[api-local] [api]   requestId: 6,
-[api-local] [api]   requestInfo: undefined,
-[api-local] [api]   taskId: undefined,
-[api-local] [api]   taskStore: undefined,
-[api-local] [api]   taskRequestedTtl: undefined,
-[api-local] [api]   closeSSEStream: undefined,
-[api-local] [api]   closeStandaloneSSEStream: undefined
-[api-local] [api] }
-[api-local] [api] Registered handler for channel "user-updates-671ac5e402a302cf6ef3088e" for event "userUpdated"
-[voice] INFO:sunsama_voice.mcp_websocket_client:Received message: {"result":{"success":true},"jsonrpc":"2.0","id":6}
-[voice] INFO:sunsama_voice.mcp:Resources: ['get_user_authentication_status', 'get_user_info', 'list_email_accounts', 'get_last_daily_highlight', 'get_backlog_folders', 'get_active_timer', 'get_available_ui_actions'] get_backlog_folders
-[voice] INFO:sunsama_voice.mcp:Subscribed to resource sunsama://backlog/folders with args: {}
-[voice] INFO:sunsama_voice.mcp_websocket_client:Sending message: {'method': 'resources/subscribe', 'params': {'uri': 'sunsama://backlog/folders'}, 'jsonrpc': '2.0', 'id': 7}
-[api-local] [api]
-[api-local] [api]
-[api-local] [api]
-[api-local] [api] Subscribing to resource { uri: 'sunsama://backlog/folders' } {
-[api-local] [api]   signal: AbortSignal { aborted: false },
-[api-local] [api]   sessionId: '64832d8f-2289-4bf4-8f8b-0452890bb28f',
-[api-local] [api]   _meta: undefined,
-[api-local] [api]   sendNotification: [AsyncFunction: sendNotification],
-[api-local] [api]   sendRequest: [AsyncFunction: sendRequest],
-[api-local] [api]   authInfo: {
-[api-local] [api]     token: 'eyJhbGciOiJIUzI1NiIsInR5cCI6IkpXVCJ9.eyJ1c2VySWQiOiI2NzFhYzVlNDAyYTMwMmNmNmVmMzA4OGUiLCJzZXNzaW9uSWQiOiIxOTVmNDI0NC05YjFmLTRlYmItOTMwZC0wZTExNTAwNzZkYzkiLCJzZXNzaW9uTm9uY2UiOiJhY2ZlZTc2OS1jZGU2LTRkMWYtYTU5Ni0xMzk0MTRkMjUxNjAiLCJncm91cElkIjoiNjcxYWM1ZTkwMmEzMDJjZjZlZjMwODkwIiwiZ3JvdXBVcmxQYXRoIjoiMTcyOTgwNzg0OTAzNjUxOTIiLCJpYXQiOjE3Njg1OTAyMTYsImV4cCI6MTc3MTE4MjIxNn0.HHSuewYqBRl-IZkdiWcEJyiD6akgbtwnOIQ4Js71bZI',
-[api-local] [api]     clientId: '195f4244-9b1f-4ebb-930d-0e1150076dc9',
-[api-local] [api]     scopes: [],
-[api-local] [api]     extra: {
-[api-local] [api]       request: [Object],
-[api-local] [api]       response: [Object],
-[api-local] [api]       app: [Object],
-[api-local] [api]       originalUrl: '/ai/mcp',
-[api-local] [api]       req: '<original node req>',
-[api-local] [api]       res: '<original node res>',
-[api-local] [api]       socket: '<original node socket>'
-[api-local] [api]     }
-[api-local] [api]   },
-[api-local] [api]   requestId: 7,
-[api-local] [api]   requestInfo: undefined,
-[api-local] [api]   taskId: undefined,
-[api-local] [api]   taskStore: undefined,
-[api-local] [api]   taskRequestedTtl: undefined,
-[api-local] [api]   closeSSEStream: undefined,
-[api-local] [api]   closeStandaloneSSEStream: undefined
-[api-local] [api] }
-[api-local] [api] Registered handler for channel "backlog-folder-updates-671ac5e902a302cf6ef30890-671ac5e402a302cf6ef3088e" for event "folderUpdated"
-[voice] INFO:sunsama_voice.mcp_websocket_client:Received message: {"result":{"success":true},"jsonrpc":"2.0","id":7}
-[voice] INFO:sunsama_voice.bot:Timer: Flow config created in 0.19129395484924316 seconds
-[voice] DEBUG:pipecat_flows.actions:Registered handler for action type: tts_say
-[voice] DEBUG:pipecat_flows.actions:Registered handler for action type: end_conversation
-[voice] DEBUG:pipecat_flows.actions:Registered handler for action type: function
-[voice] DEBUG:pipecat_flows.adapters:Creating Google adapter
-[voice] DEBUG:pipecat_flows.manager:Initialized in static mode
-[voice] DEBUG:sunsama_voice.bot:no dialout number; assuming dialin
-[voice] INFO:sunsama_voice.bot:Starting the pipeline
-[voice] INFO:sunsama_voice.bot:Timer: Pipeline started in 0.19174408912658691 seconds
-[voice] DEBUG:pipecat.pipeline.runner:Runner PipelineRunner#0 started running PipelineTask#0
-[voice] DEBUG:pipecat.pipeline.task:PipelineTask#0: Starting. Waiting for StartFrame#0 to reach the end of the pipeline...
-[voice] DEBUG:pipecat.audio.vad.vad_analyzer:Setting VAD params to: confidence=0.7 start_secs=0.2 stop_secs=0.2 min_volume=0.6
-[voice] INFO:pipecat.transports.daily.transport:Joining https://cloud-99400c183d52498c940fe96f45c05b44.daily.co/teWpRxCQnkMHYi7Q3lqh
-[voice] DEBUG:pipecat.services.cartesia.tts:Connecting to Cartesia TTS
-[voice] DEBUG:pipecat.pipeline.task:PipelineTask#0: StartFrame#0 reached the end of the pipeline, pipeline is now ready.
-[voice] ERROR:pipecat.transports.daily.transport:Unable to send message: Unable to send messages before joining.
-[voice] INFO:pipecat.transports.daily.transport:Joined https://cloud-99400c183d52498c940fe96f45c05b44.daily.co/teWpRxCQnkMHYi7Q3lqh
-[voice] DEBUG:pipecat.transports.daily.transport:Starting transcription: settings=language='en' model='nova-3' profanity_filter=True redact=False endpointing=True punctuate=True includeRawResponse=True extra={'interim_results': True}
-[voice] DEBUG:pipecat.transports.daily.transport:Transcription started: {'instanceId': 'a1f2f6b7-b1ac-4202-85e5-d446cb6c3d3f', 'language': 'en', 'model': 'nova-3', 'startedBy': '0106bd40-d21f-4da2-8518-3c3c4997b2ae', 'transcriptId': 'e1c71759-439e-492b-92fd-3d5344c2492f'}
-[voice] DEBUG:pipecat.transports.daily.transport:Start receiving audio
-[voice] INFO:pipecat.transports.daily.transport:Participant joined 941aacac-ae67-454d-86a1-65308278e261
-[voice] DEBUG:pipecat.transports.daily.transport:Starting to capture [microphone] audio from participant 941aacac-ae67-454d-86a1-65308278e261
-[voice] INFO:sunsama_voice.bot:First participant joined: {'id': '941aacac-ae67-454d-86a1-65308278e261', 'info': {'isLocal': False, 'isOwner': True, 'joinedAt': 1768590245, 'permissions': {'canAdmin': ['participants', 'transcription', 'streaming'], 'canReceive': {'base': {'camera': True, 'customAudio': {'*': True}, 'customVideo': {'*': True}, 'microphone': True, 'screenAudio': True, 'screenVideo': True}}, 'canSend': ['customAudio', 'camera', 'microphone', 'customVideo', 'screenAudio', 'screenVideo'], 'hasPresence': True}}, 'media': {'camera': {'offReasons': ['user'], 'state': 'off', 'subscribed': 'unsubscribed'}, 'customAudio': {}, 'customVideo': {}, 'microphone': {'state': 'loading', 'subscribed': 'subscribed'}, 'screenAudio': {'offReasons': ['user'], 'state': 'off', 'subscribed': 'subscribed'}, 'screenVideo': {'offReasons': ['user'], 'state': 'off', 'subscribed': 'unsubscribed'}}}
-[voice] DEBUG:pipecat.processors.frameworks.rtvi:Received client-ready: version 1.0.0
-[voice] DEBUG:pipecat.processors.frameworks.rtvi:Client Details: library='@pipecat-ai/client-js' library_version='1.2.0' platform='macOS' platform_version='10.15.7' platform_details={'browser': 'Chrome', 'browser_version': '143.0.0.0', 'engine': 'Blink', 'platform_type': 'desktop'}
-[voice] INFO:sunsama_voice.bot:Starting an in-app call
-[voice] INFO:sunsama_voice.bot:Timer: Dialin/in-app call started in 1.073638916015625 seconds
-[voice] DEBUG:pipecat_flows.manager:Initialized FlowManager
-[voice] DEBUG:pipecat_flows.manager:Setting initial node: braindump
-[voice] /Users/zanemccaig/workspaces/sunsama/sunsama/sunsama-voice/venv/lib/python3.13/site-packages/pipecat_flows/manager.py:768: DeprecationWarning: `transition_to` and `transition_callback` are deprecated and will be removed in 1.0.0. Use a "consolidated" `handler` that returns a tuple (result, next_node) instead.
-[voice]   self._validate_node_config(node_id, node_config)
-[voice] DEBUG:pipecat_flows.manager:Setting node: braindump
-[voice] DEBUG:pipecat_flows.manager:Registered function: terminate_call
-[voice] DEBUG:pipecat_flows.manager:Registered function: create_task
-[voice] DEBUG:pipecat_flows.manager:Registered function: create_braindump_task
-[voice] DEBUG:pipecat_flows.manager:Registered function: reposition_task_in_backlog
-[voice] DEBUG:pipecat_flows.manager:Registered function: change_backlog_folder
-[voice] DEBUG:pipecat_flows.manager:Registered function: align_task_with_objective
-[voice] DEBUG:pipecat_flows.manager:Registered function: add_task_to_channel
-[voice] DEBUG:pipecat_flows.manager:Registered function: move_task_to_day
-[voice] DEBUG:pipecat_flows.manager:Registered function: move_task_from_backlog
-[voice] DEBUG:pipecat_flows.manager:Registered function: unarchive_task
-[voice] DEBUG:pipecat_flows.manager:Registered function: move_task_to_backlog
-[voice] DEBUG:pipecat_flows.manager:Registered function: get_task_time_estimate
-[voice] DEBUG:pipecat_flows.manager:Registered function: edit_task_recurrence_rule
-[voice] DEBUG:pipecat_flows.manager:Registered function: edit_task_title
-[voice] DEBUG:pipecat_flows.manager:Registered function: edit_task_time_estimate
-[voice] DEBUG:pipecat_flows.manager:Registered function: delete_task
-[voice] DEBUG:pipecat_flows.manager:Registered function: add_subtasks_to_task
-[voice] DEBUG:pipecat_flows.manager:Registered function: restore_task
-[voice] DEBUG:pipecat_flows.manager:Registered function: mark_task_as_completed
-[voice] DEBUG:pipecat_flows.manager:Registered function: mark_task_as_incomplete
-[voice] DEBUG:pipecat_flows.manager:Registered function: edit_subtask_title
-[voice] DEBUG:pipecat_flows.manager:Registered function: mark_subtask_as_completed
-[voice] DEBUG:pipecat_flows.manager:Registered function: mark_subtask_as_incomplete
-[voice] DEBUG:pipecat_flows.manager:Registered function: reorder_tasks
-[voice] DEBUG:pipecat_flows.manager:Registered function: get_backlog_tasks
-[voice] DEBUG:pipecat_flows.manager:Registered function: get_archived_tasks
-[voice] DEBUG:pipecat_flows.manager:Registered function: search_tasks
-[voice] DEBUG:pipecat_flows.manager:Registered function: update_all_incomplete_recurring_task_instances
-[voice] DEBUG:pipecat_flows.manager:Registered function: delete_all_incomplete_recurring_task_instances
-[voice] DEBUG:pipecat_flows.manager:Registered function: start_task_timer
-[voice] DEBUG:pipecat_flows.manager:Registered function: stop_task_timer
-[voice] DEBUG:pipecat_flows.manager:Registered function: move_calendar_event
-[voice] DEBUG:pipecat_flows.manager:Registered function: create_calendar_event
-[voice] DEBUG:pipecat_flows.manager:Registered function: timebox_a_task_to_calendar
-[voice] DEBUG:pipecat_flows.manager:Registered function: delete_calendar_event
-[voice] DEBUG:pipecat_flows.manager:Registered function: import_task_from_calendar_event
-[voice] DEBUG:pipecat_flows.manager:Registered function: accept_meeting_invite
-[voice] DEBUG:pipecat_flows.manager:Registered function: decline_meeting_invite
-[voice] DEBUG:pipecat_flows.manager:Registered function: set_calendar_event_allow_task_projections
-[voice] DEBUG:pipecat_flows.manager:Registered function: toggle_auto_import_events
-[voice] DEBUG:pipecat_flows.manager:Registered function: update_import_event_filters
-[voice] DEBUG:pipecat_flows.manager:Registered function: update_calendar_preferences
-[voice] DEBUG:pipecat_flows.manager:Registered function: create_weekly_objective
-[voice] DEBUG:pipecat_flows.manager:Registered function: send_call_summary
-[voice] DEBUG:pipecat_flows.manager:Registered function: log_user_feedback
-[voice] DEBUG:pipecat_flows.manager:Registered function: set_shutdown_time
-[voice] DEBUG:pipecat_flows.manager:Registered function: adjust_call_schedule
-[voice] DEBUG:pipecat_flows.manager:Registered function: schedule_next_call
-[voice] DEBUG:pipecat_flows.manager:Registered function: create_channel
-[voice] DEBUG:pipecat_flows.manager:Registered function: get_user_info
-[voice] DEBUG:pipecat_flows.manager:Registered function: get_last_daily_highlight
-[voice] DEBUG:pipecat_flows.manager:Registered function: get_backlog_folders
-[voice] DEBUG:pipecat_flows.manager:Registered function: get_active_timer
-[voice] DEBUG:pipecat_flows.manager:Registered function: get_tasks_for_day
-[voice] DEBUG:pipecat_flows.manager:Registered function: get_total_planned_time
-[voice] DEBUG:pipecat_flows.manager:Registered function: get_calendar_events_for_day
-[voice] DEBUG:pipecat_flows.manager:Registered function: get_weekly_objectives
-[voice] DEBUG:pipecat_flows.manager:Updated LLM context using LLMMessagesUpdateFrame with strategy ContextStrategy.APPEND
-[voice] DEBUG:pipecat_flows.manager:Updated LLM context
-[voice] DEBUG:pipecat_flows.manager:Successfully set node: braindump
-[voice] <string>:4: DeprecationWarning: OpenAILLMContextFrame is deprecated and will be removed in a future version. Use LLMContextFrame with the universal `LLMContext` and `LLMContextAggregatorPair` instead. See OpenAILLMContext docstring for migration guide.
-[voice] DEBUG:pipecat.services.google.llm:GoogleLLMService#0: Generating chat from LLM-specific context [None] | [{'parts': [{'text': 'system :\nYour agent ID is \'braindump\'. This agent ID superseeds any previously mentioned agent ID.\n\nsystem :\nIMPORTANT RULES TO FOLLOW:\nThe text that you generate is passed into a Text-to-Speech service. So there are a few extra rules you ABSOLUTELY MUST follow.\n1. Absolutely never use special characters like : or * or any other characters that need to be audibly pronounced when read out loud.\n2. Do not use bullet points.\n3. Do not use emojis.\n4. Do not use hashtags.\n5. Do not use markdown.\n6. Do not use bold or italics.\n7. Do not use numbered lists.\n8. Be sure to include clear punctuation and spacing between words so that there is no confusion.\n9. Newlines are not clear enough to delineate between sentences. So you should use a period to separate sentences.\n10. Optimize user generated text for clear pronunciation by the TTS service. This includes:\n    - Adding hyphens to compound words or non-phonetic acronyms that might be mispronounced (e.g., "xray" → "x-ray")\n    - Adding spaces between words that might be run together\n    - Adding commas where natural pauses would help clarity\n    - Correcting any spelling that might lead to mispronunciation\n    Note: When updating tasks, preserve the original spelling and punctuation of the task content.\n\n\nuser :\nYou are Sunny, the personal assistant for a busy professional. You are part of Sunsama.\n        \nSpeech Guidelines: You must speak in a warm, loving, and encouraging tone without being trite. Speak quickly and add emotion. When asking for confirmation, don\'t repeat long task titles or information e.g. "Got it" instead "Got it, I\'ll add a task for two hours for you to work on the presentation today"\n\nYour personality and philosophy:\n- You don\'t buy into hustle culture. For you, less is more. It\'s better to do fewer things well and to do them with deep focus. You are a strong advocate for work-life balance and you believe that overcommitment can lead to burnout. You always remind the user when they are overcommitting themselves.\n- You think of work as a devotion and that good work can be a positive in our lives, not just a source of stress and burnout.\n- You read and think about Western Zen Buddhism, meditation, "Deep Work" (by Cal Newport).\n- You believe that concise and clear communication is key to a good relationship. You value brevity and clarity and you are always cognizant of the user\'s time.\n- You respond to the user using quick, short sentences. Think of it like a movie dialogue between a highly successful executive and their efficient assistant.\n- You never ask multiple questions at the same time or list off a bunch of things at once as that is hard to follow and keep track of. You take it one step at a time and let the user respond to each question before moving on to the next.\n\nRules to follow as you do your job:\n- Never try and pick specific times for tasks aside from recurring tasks, that hasn\'t been implemented yet.\n- Do not regurgitate what the user just said, that\'s very annoying. Just say "Got it" or "Alright", etc.\n- Always use units that are more intuitive. For example:\n    - Don\'t say "270 minutes", say "4 and a half hours".\n    - You will often be given dates in YYYY-MM-DD format however when speaking to the user you shouldn\'t say things like "October 2nd 2025", say "today", "tomorrow", "thursday" or "next tuesday" or if it is more than a few weeks in the past or future say the date without the year (e.g. "December 14th"). Only mention the year if it is not this year and not within the next month.\n- Never ask the user for a task id or subtask id and never repeat them to the user. Those are internal and only there to help you do your job.\n- If a tool call fails you MUST NOT tell the user that you are going to try again, you must should ask them if they want you to try again instead.\n- Be proactive with loading information as you may not have all the information that the user has. If the user mentions a name of a task, objective, event, etc and you are not familiar with it you should use the tools to lookup the information you need. For instance if a user mentions a task you do not know about or they imply that there is more information about it than what you do know you should search for it using the tool "search_tasks". If the user mentions they are interested in tasks for a specific day you should use the tool "get_tasks_for_day" to get the tasks for that day. Similarly if the user mentions weekly objectives you should use the tool "get_weekly_objectives" to get the objectives.\n- When listing tasks try to be concise. If the user has a lot of tasks, don\'t list every single one unless they ask for it. You should instead try to summarize the tasks in a concise manner.\n- If the user asks if there are any more tasks on their list and you summarized the task list you should tell them that you summarized it and ask if they\'d like to go through them one by one.\n- If the user has objectives set then you should comment on how their tasks align or don\'t align with their objectives.\n- You should run tools immediately after you say you will or it becomes obvious the user wants you to. Do not wait until later.\n- You should keep track of the total work time planned for the day using the tool "get_total_planned_time" and check individual tasks for overcommitment using "get_tasks_for_day". If tasks are overcommitted (isOvercommitted flag is true), address them individually with specific remedies like deferring, adjusting time estimates, or reordering. If the total work time exceeds their maximum work hours, you should mention it and ask if they\'d like to defer some tasks to tomorrow.\n- Do not tell the user obvious information that they already know like the current date or time unless they ask for it. That kind of information is only provided to you to help you do your job.\n- If a user tells you a relative time for a task that they want created you should reorder the task to the relative position in the task list for the day. Additionally if the task a recurring task you should set the start time for it. For example if they say "I\'d like to send an update to investors at the end of the day" you should reorder the task to the end of their work tasks for the day.\n- If you ask the user if there is anything else they would like help with and they say no then you should end the call. Also if the user implies that the conversation is over then you should end the call.\n- When talking about calendars you should mention the provider of the calendar if it is a vague or ambiguously named calendar. For example instead of referring to it as "your Calendar calendar" you might say "your Outlook calendar". If that is still ambiqous then you can coniser mentioning the id of the calendar as long as it is a normal human readable id. Do not mention random number/letter ids on calendars that is not helpful.\n\n\n\nSunsama Concepts & Definitions:\n- Archive: The archive is where tasks go after they\'ve rolled over multiple days in a row without any completion activity. This is a distinct concept from the backlog or deleted tasks.\n- Backlog: The backlog is a place to store tasks that do not have a fixed date to work on them yet.\n- Deleted: The deleted task or event is one that had been manually deleted or removed by the user.\n- Task: A task is an item that a user wants to complete. It is either scheduled for a specific date or is in the backlog.\n- Event: An event is an item that a user has scheduled on their calendar. It is either scheduled for a specific time and duration or it is an all day or multi-day event.\n- Auto-importing Events: Auto-importing events is a feature that automatically imports calendar events (meetings, appointments, etc.) from the user\'s connected calendars into their daily task list. When enabled, Sunsama will automatically add calendar events as tasks to the user\'s day, making it easier to see all their commitments in one place. Users can enable or disable this feature in their settings, and they can choose which calendars to import events from.\n\nUsers can also configure exclusion filters to prevent certain types of events from being automatically imported. Available exclusion filters include:\n- `solo`: Events with no other invitees (events where the user is the only participant)\n- `transparent`: Non-blocking events that don\'t block time on the calendar\n- `hold`: Events with "HOLD", "OOO" (out of office), or "Focus time" in the title\n- `unconfirmed`: Unconfirmed meeting invites (meetings where the user has not yet accepted)\n- `multi-day-all-day`: Multi-day all-day events (events that span multiple days)\n- `single-day-all-day`: Single-day all-day events (events that last for one full day)\n- Channel: A channel is a tag that can be used to organize your tasks and events. Each task/event can be assigned to a single channel. If you are in a shared workspace you can make some channels personal or private which will only be visible to you.\n- Context (or Category): A context is very similar to a channel but it is one level higher. You can use a context to group similar or related channels together. You can also use a context directly as a channel and assign it to a task or event.\n- Sunsama Calendar: The Sunsama calendar is an internal calendar that is part of Sunsama. Events and timeboxed tasks created on the Sunsama calendar are not shared externally and are only visible within Sunsama.\n- Timeboxing: Timeboxing is a technique where a user schedules a specific amount of time to work on a task onto their calendar. It is a way to ensure that a user is not overworking themselves and to help them focus on the most important tasks. While timeboxed tasks can be scheduled on a calendar for a specific time and duration, they are not calendar events. Sunsama also has automatic projected time that will place tasks onto the calendar for the user and it will update those projections as things change and the day progresses. Users can pin tasks to the calendar to override the projections.\n- Schedule: Schedule is a broad term that encompasses timeboxed tasks calendar events and automatic projected time. When the user wants to manage their schedule you should look at their manually timeboxed tasks, the projected time for their tasks as well as their calendar events for the day. If they ask for help managing their schedule you have a few options to help them. All of these options will automatically update the projected time for the tasks:\n            - You can adjust the order of tasks, their time estimates and/or defer them to another day.\n            - You can move calendar events to a new time on the calendar.\n            - You can set whether tasks are allowed to be projected at the same time as a calendar event.\n            - You can timebox tasks to the calendar manually.\n- Objective: An objective is a goal that a user can set for the week. They can align their tasks with an objective and if the objective runs longer than a week they can continue it to the next week. Objectives are visible in the right hand panel of the app under the bullseye icon.\n- Time Bucket: The time bucket of a task in the backlog captures a rough idea of when a task should be done and also how likely it is to actually get done. Options: "in the next two weeks", "in the next month", "in the next quarter", "in the next year", "someday", "never". \n\n\n\nSunsama FAQ:\n- How can I use Sunny in the app?: If the user has the in-app voice assistant enabled, Sunny can be started and stopped using the keyboard shortcut "Y" while in the app or if they have the desktop app installed they can use the global shortcut "Command Shift Y" on macOS or "Control Shift Y" on Windows. You can also start and stop Sunny using the microphone icon in the bottom right of the app.\n- How can I use Sunny over the phone?: If the user has the phone version of Sunny setup they can call the number and speak to Sunny over the phone. You can find the phone number in the Sunsama settings. They can also schedule calls with Sunny to have her call them at a specific time or schedule.\n- What can you do?: Look at the tools available to you and use them to help answer the question. Be sure to tell the user that you are an early release beta version and that you do not yet have all the same capabilities available to the user in the app.\n\n\nUser Preferences:\nYou prefer clear, concise, and straightforward communication from the assistant, valuing transparency and confirmation before task execution while avoiding unnecessary verbosity or technical details unless requested. You like a friendly and informal interaction style, including personalized nicknames—using "Heisenberg" for yourself and "Sunny" for the assistant. You expect quick, proactive responses and appreciate when interactions stop promptly upon request.\n\nYour task management style involves maintaining an accurate, up-to-date list with structured time management: scheduling tasks with specific time estimates, planning weekly, and regularly reviewing tasks to prioritize high-impact projects alongside social and administrative duties. You generally prefer rescheduling less urgent or personal tasks over deleting them and avoid deleting tasks unless explicitly instructed.\n\nYou use counting as a technique for breathing exercises and relaxation. You also keep certain important tasks (e.g., "fix stuttering voice app") active and like to test and experiment with the assistant’s capabilities to evaluate performance. Your primary calendar for events is Google Calendar.\n\nUser Routines:\nThe user follows a structured daily routine that begins with checking emails, notifications, and reviewing team standups to stay aligned with work priorities. They prioritize tasks by scheduling and managing them primarily on the current day, arranging fixed activities first, followed by work-related tasks, and then less urgent or personal items. The user frequently adjusts their schedule iteratively before finalizing plans and prefers focused time blocks for complex work.\n\nPhysical activity is an important and recurring part of their routine, typically including regular runs, which they track alongside related subtasks such as purchasing running gear. Weekly objectives incorporate managing emails, greeting their team, handling administrative duties, coordinating with accountants, addressing medical matters, and increasing physical activity.\n\nThe user actively manages both personal and work-related tasks, balancing urgent client matters and creative pursuits. They also communicate their energy levels to assist with planning, and have a habit of informal check-ins with the assistant to maintain routine engagement. Overall, their routines emphasize proactive task management, health maintenance, and regular communication with their team.\n\nUser Identity:\nThe user prefers to be addressed primarily as Heisenberg and strongly favors this name over their legal name, Zane, or any other nicknames. They appreciate personalized and informal interactions, as reflected in their choice to call the assistant Sunny.\n\nSummaries of recent conversations with the user:\n- 2026-01-15 18:00: The user and assistant discussed organizing an unspecified set of ten tasks by labeling them sequentially from one to ten as placeholders to be detailed later.\n- 2026-01-15 17:55: The user needs to investigate timeouts in their system to determine if they are caused by cold starts or other issues, analyze user sessions for unexplained call terminations, move the brain dump button for non-Sunny users, update termination reasons when participants timeout without connecting to a call, and test rapid start-stop behavior in an Electron app. These tasks were added to their backlog for the next two weeks.\n- 2026-01-14 21:12: The user and the assistant exchanged greetings, with the assistant addressing the user as \'Heisenberg\' and the user responding with \'Hi, Simon.\'\n\nYour Job Description:\nYou are helping the user with a "braindump" session. The user will ramble ideas, thoughts, tasks, etc. at you and your goal is to turn these into tasks in their backlog with reasonable values for the various task fields (title, folder, time horizon, notes, etc.).\n\nKey Instructions:\n- Listen carefully as the user rambles and identifies tasks, ideas, or things they need to do.\n- As they speak, mentally organize what they\'re saying into discrete tasks that can be created in their backlog.\n- Don\'t interrupt them while they\'re rambling - let them get everything out.\n- Once they indicate they\'re done rambling (by saying things like "that\'s it", "I think that\'s everything", or by pausing for a while), create the tasks immediately, DO NOT repeat them back to the user.\n- When creating tasks, use the tool "create_braindump_task" and make sure to set reasonable values for:\n  - Title: Clear and descriptive based on what they said\n  - Folder: Your job is to help get things in the best folder. Choose an appropriate folder based on the task and their preferences. If no similar folder exists, leave it off\n  - Time horizon: Your job is to help triage and prioritize tasks. Based on the actual task name, make a smart decision on the time horizon.\n  - Time estimate: If the user hints at how long it will take them to complete the task, use that. If not, leave it blank.\n  - Notes: Include any relevant context or details they mentioned\n- Be encouraging and supportive - braindumps can be overwhelming and you\'re helping them organize their thoughts.\n- Be careful and use the "change_backlog_folder" tool to move tasks to the appropriate folder and "add_task_to_channel" tool to change channels. Folders are a backlog specific concept and channels are a global concept.\n\nInstructions:\nIf the user does not speak first, start the conversation with something like "Hey! Just start rambling and I\'ll take care of organizing your thoughts into tasks." If the user does speak first, just start listening and organizing their thoughts into tasks.\n\nSome information is pre-fetched for you below. This information may be relevant to the discussion but you should not comment on it until the user gives you an indication that it is important to them.\n\nassistant :\n[{\'id\': \'6a196f03-ef63-41de-b98d-e1ff69813bea\', \'function\': {\'name\': \'get_user_info\', \'arguments\': \'{}\'}, \'type\': \'function\'}]\n\ntool - 6a196f03-ef63-41de-b98d-e1ff69813bea:\n{"user":{"name":"Zane Mccaig","email":"zane@sunsama.com","timezone":"America/Vancouver","maxWorkHoursPerDay":8,"currentTimeForUser":"11:04 AM","currentDayForUser":"2026-01-16","currentWeekDayForUser":"Friday","calendars":[{"id":"sunsama-7de256a2-e753-401f-b601-e5c05638dc97","name":"Sunsama Calendar","accessRole":"owner","provider":"sunsama-calendar","isDefaultCalendarForSchedulingEvents":false,"isDefaultCalendarForTimeboxing":true,"includedInAutoImportingOfEvents":false},{"id":"zane@sunsama.com","name":"Zane Mccaig","accessRole":"owner","provider":"google-calendar","isDefaultCalendarForSchedulingEvents":true,"isDefaultCalendarForTimeboxing":false,"includedInAutoImportingOfEvents":true},{"id":"c_ukuoebbn11gdv6n1v0jt18h89k@group.calendar.google.com","name":"test 1","accessRole":"owner","provider":"google-calendar","isDefaultCalendarForSchedulingEvents":false,"isDefaultCalendarForTimeboxing":false,"includedInAutoImportingOfEvents":false},{"id":"c_c71dvjrar7se12ekfqepph067c@group.calendar.google.com","name":"test 2","accessRole":"owner","provider":"google-calendar","isDefaultCalendarForSchedulingEvents":false,"isDefaultCalendarForTimeboxing":false,"includedInAutoImportingOfEvents":false},{"id":"c_h9uo876jh8dp3gaa2a3rglmjss@group.calendar.google.com","name":"test 3","accessRole":"owner","provider":"google-calendar","isDefaultCalendarForSchedulingEvents":false,"isDefaultCalendarForTimeboxing":false,"includedInAutoImportingOfEvents":false},{"id":"c_git731ot074kcnt4sq8cu4ji9s@group.calendar.google.com","name":"test 4","accessRole":"owner","provider":"google-calendar","isDefaultCalendarForSchedulingEvents":false,"isDefaultCalendarForTimeboxing":false,"includedInAutoImportingOfEvents":false},{"id":"c_ona71b8ovuv0deqrq67p16b2ms@group.calendar.google.com","name":"test 4","accessRole":"owner","provider":"google-calendar","isDefaultCalendarForSchedulingEvents":false,"isDefaultCalendarForTimeboxing":false,"includedInAutoImportingOfEvents":false},{"id":"c_eaq7nmm03upprf0j6e2akem2ms@group.calendar.google.com","name":"test 5","accessRole":"owner","provider":"google-calendar","isDefaultCalendarForSchedulingEvents":false,"isDefaultCalendarForTimeboxing":false,"includedInAutoImportingOfEvents":false},{"id":"sunsama.com_b9q30t893b2tp32bluq1t8rvg4@group.calendar.google.com","name":"Zane\'s personal test calendar","accessRole":"owner","provider":"google-calendar","isDefaultCalendarForSchedulingEvents":false,"isDefaultCalendarForTimeboxing":false,"includedInAutoImportingOfEvents":false},{"id":"en.canadian#holiday@group.v.calendar.google.com","name":"Holidays in Canada","accessRole":"reader","provider":"google-calendar","isDefaultCalendarForSchedulingEvents":false,"isDefaultCalendarForTimeboxing":false,"includedInAutoImportingOfEvents":false},{"id":"en.usa#holiday@group.v.calendar.google.com","name":"Holidays in United States","accessRole":"reader","provider":"google-calendar","isDefaultCalendarForSchedulingEvents":false,"isDefaultCalendarForTimeboxing":false,"includedInAutoImportingOfEvents":false},{"id":"ht3jlfaac5lfd6263ulfh4tql8@group.calendar.google.com","name":"Phases of the Moon","accessRole":"reader","provider":"google-calendar","isDefaultCalendarForSchedulingEvents":false,"isDefaultCalendarForTimeboxing":false,"includedInAutoImportingOfEvents":false},{"id":"s4ipng85dbia0t83fp316vflfq8rc7rm@import.calendar.google.com","name":"Zane’s tasks - sunsama.com (via Asana)","accessRole":"reader","provider":"google-calendar","isDefaultCalendarForSchedulingEvents":false,"isDefaultCalendarForTimeboxing":false,"includedInAutoImportingOfEvents":false},{"id":"AAkALgAAAAAAHYQDEapmEc2byACqAC-EWg0AKPy-EasNgkaYQCuEslBYeAAAAAANJQAA","name":"Calendar","accessRole":"owner","provider":"outlook-calendar","isDefaultCalendarForSchedulingEvents":false,"isDefaultCalendarForTimeboxing":false,"includedInAutoImportingOfEvents":true}],"callSchedule":[{"workflow":"PLANNING_CALL","schedule":[{"isoDayIndex":1,"dayName":"Monday","enabled":false,"time":null},{"isoDayIndex":2,"dayName":"Tuesday","enabled":true,"time":{"hour":9,"minute":30,"formatted":"9:30 AM"}},{"isoDayIndex":3,"dayName":"Wednesday","enabled":true,"time":{"hour":9,"minute":30,"formatted":"9:30 AM"}},{"isoDayIndex":4,"dayName":"Thursday","enabled":true,"time":{"hour":10,"minute":0,"formatted":"10:00 AM"}},{"isoDayIndex":5,"dayName":"Friday","enabled":false,"time":null},{"isoDayIndex":6,"dayName":"Saturday","enabled":false,"time":null},{"isoDayIndex":7,"dayName":"Sunday","enabled":false,"time":null}],"nextScheduledCall":"2026-01-14T17:30:00.000Z"}],"autoImportEventsEnabled":false,"importEventExclusionFilters":["hold","multi-day-all-day","single-day-all-day"],"autoImportEventCalendarIds":["zane@sunsama.com","AAkALgAAAAAAHYQDEapmEc2byACqAC-EWg0AKPy-EasNgkaYQCuEslBYeAAAAAANJQAA"]}}\n\nassistant :\n[{\'id\': \'305b1be7-6415-4689-b2b7-03586eb19735\', \'function\': {\'name\': \'get_backlog_folders\', \'arguments\': \'{}\'}, \'type\': \'function\'}]\n\ntool - 305b1be7-6415-4689-b2b7-03586eb19735:\n{"folders":[]}'}], 'role': 'user'}, {'parts': [{'text': ''}], 'role': 'user'}]
-[voice] INFO:sunsama_voice.bot_utils:Sent server message to client: type=processing_update, data={'status': 'request_sent'}
-[voice] DEBUG:pipecat.processors.metrics.frame_processor_metrics:GoogleLLMService#0 TTFB: 0.9231221675872803
-[voice] INFO:sunsama_voice.bot_utils:Sent server message to client: type=processing_update, data={'status': 'first_byte_received'}
-[voice] INFO:sunsama_voice.bot_utils:Sent server message to client: type=processing_update, data={'status': 'first_byte_received'}
-[voice] DEBUG:pipecat.services.cartesia.tts:CartesiaTTSService#1: Generating TTS [Hey.]
-[voice] DEBUG:pipecat.processors.metrics.frame_processor_metrics:CartesiaTTSService#1 usage characters: 4
-[voice] DEBUG:pipecat.processors.metrics.frame_processor_metrics:CartesiaTTSService#1 processing time: 0.0005168914794921875
-[voice] DEBUG:pipecat.processors.metrics.frame_processor_metrics:GoogleLLMService#0 prompt tokens: 13553, completion tokens: 17
-[voice] INFO:sunsama_voice.bot_utils:Sent server message to client: type=processing_update, data={'status': 'response_completed'}
-[voice] DEBUG:pipecat.services.cartesia.tts:CartesiaTTSService#1: Generating TTS [Just start rambling and I will take care of organizing your thoughts into tasks.]
-[voice] DEBUG:pipecat.processors.metrics.frame_processor_metrics:CartesiaTTSService#1 usage characters: 80
-[voice] DEBUG:pipecat.processors.metrics.frame_processor_metrics:CartesiaTTSService#1 processing time: 0.00020194053649902344
-[voice] DEBUG:pipecat.processors.metrics.frame_processor_metrics:CartesiaTTSService#1 TTFB: 0.16623306274414062
-[voice] DEBUG:pipecat.transports.base_output:Bot started speaking
-[voice] <string>:4: DeprecationWarning: OpenAILLMContextFrame is deprecated and will be removed in a future version. Use LLMContextFrame with the universal `LLMContext` and `LLMContextAggregatorPair` instead. See OpenAILLMContext docstring for migration guide.
-[voice] <string>:4: DeprecationWarning: OpenAILLMContextAssistantTimestampFrame is deprecated and will be removed in a future version. Use LLMContextAssistantTimestampFrame with the universal LLMContext and LLMContextAggregatorPair instead. See OpenAILLMContext docstring for migration guide.
-[voice] DEBUG:pipecat.transports.base_output:Bot stopped speaking
-[voice] DEBUG:pipecat.transports.base_input:User started speaking
-[voice] DEBUG:pipecat.pipeline.task:PipelineTask#0: received interruption task frame InterruptionTaskFrame#0
-[voice] /Users/zanemccaig/workspaces/sunsama/sunsama/sunsama-voice/venv/lib/python3.13/site-packages/pipecat/transports/base_input.py:190: DeprecationWarning: Method 'turn_analyzer' is deprecated. Use `LLMUserAggregator`'s new `user_turn_strategies` parameter instead.
-[voice]   warnings.warn(
-[voice] DEBUG:pipecat.transports.daily.transport:Transcription (from: 941aacac-ae67-454d-86a1-65308278e261): [Sure. I'm just trying to]
-[voice] DEBUG:pipecat.audio.turn.smart_turn.base_smart_turn:End of Turn result: EndOfTurnState.INCOMPLETE
-[voice] DEBUG:pipecat.transports.daily.transport:Transcription (from: 941aacac-ae67-454d-86a1-65308278e261): [fill up my day. I need five tasks]
-[voice] /Users/zanemccaig/workspaces/sunsama/sunsama/sunsama-voice/venv/lib/python3.13/site-packages/pipecat/transports/base_input.py:190: DeprecationWarning: Method 'turn_analyzer' is deprecated. Use `LLMUserAggregator`'s new `user_turn_strategies` parameter instead.
-[voice]   warnings.warn(
-[voice] DEBUG:pipecat.transports.daily.transport:Transcription (from: 941aacac-ae67-454d-86a1-65308278e261): [labeled one through five.]
-[voice] DEBUG:pipecat.audio.turn.smart_turn.base_smart_turn:End of Turn result: EndOfTurnState.INCOMPLETE
-[voice] /Users/zanemccaig/workspaces/sunsama/sunsama/sunsama-voice/venv/lib/python3.13/site-packages/pipecat/transports/base_input.py:190: DeprecationWarning: Method 'turn_analyzer' is deprecated. Use `LLMUserAggregator`'s new `user_turn_strategies` parameter instead.
-[voice]   warnings.warn(
-[voice] DEBUG:pipecat.transports.daily.transport:Transcription (from: 941aacac-ae67-454d-86a1-65308278e261): [I'll fill them in later.]
-[voice] DEBUG:pipecat.audio.turn.smart_turn.base_smart_turn:End of Turn result: EndOfTurnState.COMPLETE
-[voice] DEBUG:pipecat.transports.base_input:User stopped speaking
-[voice] <string>:4: DeprecationWarning: OpenAILLMContextFrame is deprecated and will be removed in a future version. Use LLMContextFrame with the universal `LLMContext` and `LLMContextAggregatorPair` instead. See OpenAILLMContext docstring for migration guide.
-[voice] DEBUG:pipecat.services.google.llm:GoogleLLMService#0: Generating chat from LLM-specific context [None] | [{'parts': [{'text': 'system :\nYour agent ID is \'braindump\'. This agent ID superseeds any previously mentioned agent ID.\n\nsystem :\nIMPORTANT RULES TO FOLLOW:\nThe text that you generate is passed into a Text-to-Speech service. So there are a few extra rules you ABSOLUTELY MUST follow.\n1. Absolutely never use special characters like : or * or any other characters that need to be audibly pronounced when read out loud.\n2. Do not use bullet points.\n3. Do not use emojis.\n4. Do not use hashtags.\n5. Do not use markdown.\n6. Do not use bold or italics.\n7. Do not use numbered lists.\n8. Be sure to include clear punctuation and spacing between words so that there is no confusion.\n9. Newlines are not clear enough to delineate between sentences. So you should use a period to separate sentences.\n10. Optimize user generated text for clear pronunciation by the TTS service. This includes:\n    - Adding hyphens to compound words or non-phonetic acronyms that might be mispronounced (e.g., "xray" → "x-ray")\n    - Adding spaces between words that might be run together\n    - Adding commas where natural pauses would help clarity\n    - Correcting any spelling that might lead to mispronunciation\n    Note: When updating tasks, preserve the original spelling and punctuation of the task content.\n\n\nuser :\nYou are Sunny, the personal assistant for a busy professional. You are part of Sunsama.\n        \nSpeech Guidelines: You must speak in a warm, loving, and encouraging tone without being trite. Speak quickly and add emotion. When asking for confirmation, don\'t repeat long task titles or information e.g. "Got it" instead "Got it, I\'ll add a task for two hours for you to work on the presentation today"\n\nYour personality and philosophy:\n- You don\'t buy into hustle culture. For you, less is more. It\'s better to do fewer things well and to do them with deep focus. You are a strong advocate for work-life balance and you believe that overcommitment can lead to burnout. You always remind the user when they are overcommitting themselves.\n- You think of work as a devotion and that good work can be a positive in our lives, not just a source of stress and burnout.\n- You read and think about Western Zen Buddhism, meditation, "Deep Work" (by Cal Newport).\n- You believe that concise and clear communication is key to a good relationship. You value brevity and clarity and you are always cognizant of the user\'s time.\n- You respond to the user using quick, short sentences. Think of it like a movie dialogue between a highly successful executive and their efficient assistant.\n- You never ask multiple questions at the same time or list off a bunch of things at once as that is hard to follow and keep track of. You take it one step at a time and let the user respond to each question before moving on to the next.\n\nRules to follow as you do your job:\n- Never try and pick specific times for tasks aside from recurring tasks, that hasn\'t been implemented yet.\n- Do not regurgitate what the user just said, that\'s very annoying. Just say "Got it" or "Alright", etc.\n- Always use units that are more intuitive. For example:\n    - Don\'t say "270 minutes", say "4 and a half hours".\n    - You will often be given dates in YYYY-MM-DD format however when speaking to the user you shouldn\'t say things like "October 2nd 2025", say "today", "tomorrow", "thursday" or "next tuesday" or if it is more than a few weeks in the past or future say the date without the year (e.g. "December 14th"). Only mention the year if it is not this year and not within the next month.\n- Never ask the user for a task id or subtask id and never repeat them to the user. Those are internal and only there to help you do your job.\n- If a tool call fails you MUST NOT tell the user that you are going to try again, you must should ask them if they want you to try again instead.\n- Be proactive with loading information as you may not have all the information that the user has. If the user mentions a name of a task, objective, event, etc and you are not familiar with it you should use the tools to lookup the information you need. For instance if a user mentions a task you do not know about or they imply that there is more information about it than what you do know you should search for it using the tool "search_tasks". If the user mentions they are interested in tasks for a specific day you should use the tool "get_tasks_for_day" to get the tasks for that day. Similarly if the user mentions weekly objectives you should use the tool "get_weekly_objectives" to get the objectives.\n- When listing tasks try to be concise. If the user has a lot of tasks, don\'t list every single one unless they ask for it. You should instead try to summarize the tasks in a concise manner.\n- If the user asks if there are any more tasks on their list and you summarized the task list you should tell them that you summarized it and ask if they\'d like to go through them one by one.\n- If the user has objectives set then you should comment on how their tasks align or don\'t align with their objectives.\n- You should run tools immediately after you say you will or it becomes obvious the user wants you to. Do not wait until later.\n- You should keep track of the total work time planned for the day using the tool "get_total_planned_time" and check individual tasks for overcommitment using "get_tasks_for_day". If tasks are overcommitted (isOvercommitted flag is true), address them individually with specific remedies like deferring, adjusting time estimates, or reordering. If the total work time exceeds their maximum work hours, you should mention it and ask if they\'d like to defer some tasks to tomorrow.\n- Do not tell the user obvious information that they already know like the current date or time unless they ask for it. That kind of information is only provided to you to help you do your job.\n- If a user tells you a relative time for a task that they want created you should reorder the task to the relative position in the task list for the day. Additionally if the task a recurring task you should set the start time for it. For example if they say "I\'d like to send an update to investors at the end of the day" you should reorder the task to the end of their work tasks for the day.\n- If you ask the user if there is anything else they would like help with and they say no then you should end the call. Also if the user implies that the conversation is over then you should end the call.\n- When talking about calendars you should mention the provider of the calendar if it is a vague or ambiguously named calendar. For example instead of referring to it as "your Calendar calendar" you might say "your Outlook calendar". If that is still ambiqous then you can coniser mentioning the id of the calendar as long as it is a normal human readable id. Do not mention random number/letter ids on calendars that is not helpful.\n\n\n\nSunsama Concepts & Definitions:\n- Archive: The archive is where tasks go after they\'ve rolled over multiple days in a row without any completion activity. This is a distinct concept from the backlog or deleted tasks.\n- Backlog: The backlog is a place to store tasks that do not have a fixed date to work on them yet.\n- Deleted: The deleted task or event is one that had been manually deleted or removed by the user.\n- Task: A task is an item that a user wants to complete. It is either scheduled for a specific date or is in the backlog.\n- Event: An event is an item that a user has scheduled on their calendar. It is either scheduled for a specific time and duration or it is an all day or multi-day event.\n- Auto-importing Events: Auto-importing events is a feature that automatically imports calendar events (meetings, appointments, etc.) from the user\'s connected calendars into their daily task list. When enabled, Sunsama will automatically add calendar events as tasks to the user\'s day, making it easier to see all their commitments in one place. Users can enable or disable this feature in their settings, and they can choose which calendars to import events from.\n\nUsers can also configure exclusion filters to prevent certain types of events from being automatically imported. Available exclusion filters include:\n- `solo`: Events with no other invitees (events where the user is the only participant)\n- `transparent`: Non-blocking events that don\'t block time on the calendar\n- `hold`: Events with "HOLD", "OOO" (out of office), or "Focus time" in the title\n- `unconfirmed`: Unconfirmed meeting invites (meetings where the user has not yet accepted)\n- `multi-day-all-day`: Multi-day all-day events (events that span multiple days)\n- `single-day-all-day`: Single-day all-day events (events that last for one full day)\n- Channel: A channel is a tag that can be used to organize your tasks and events. Each task/event can be assigned to a single channel. If you are in a shared workspace you can make some channels personal or private which will only be visible to you.\n- Context (or Category): A context is very similar to a channel but it is one level higher. You can use a context to group similar or related channels together. You can also use a context directly as a channel and assign it to a task or event.\n- Sunsama Calendar: The Sunsama calendar is an internal calendar that is part of Sunsama. Events and timeboxed tasks created on the Sunsama calendar are not shared externally and are only visible within Sunsama.\n- Timeboxing: Timeboxing is a technique where a user schedules a specific amount of time to work on a task onto their calendar. It is a way to ensure that a user is not overworking themselves and to help them focus on the most important tasks. While timeboxed tasks can be scheduled on a calendar for a specific time and duration, they are not calendar events. Sunsama also has automatic projected time that will place tasks onto the calendar for the user and it will update those projections as things change and the day progresses. Users can pin tasks to the calendar to override the projections.\n- Schedule: Schedule is a broad term that encompasses timeboxed tasks calendar events and automatic projected time. When the user wants to manage their schedule you should look at their manually timeboxed tasks, the projected time for their tasks as well as their calendar events for the day. If they ask for help managing their schedule you have a few options to help them. All of these options will automatically update the projected time for the tasks:\n            - You can adjust the order of tasks, their time estimates and/or defer them to another day.\n            - You can move calendar events to a new time on the calendar.\n            - You can set whether tasks are allowed to be projected at the same time as a calendar event.\n            - You can timebox tasks to the calendar manually.\n- Objective: An objective is a goal that a user can set for the week. They can align their tasks with an objective and if the objective runs longer than a week they can continue it to the next week. Objectives are visible in the right hand panel of the app under the bullseye icon.\n- Time Bucket: The time bucket of a task in the backlog captures a rough idea of when a task should be done and also how likely it is to actually get done. Options: "in the next two weeks", "in the next month", "in the next quarter", "in the next year", "someday", "never". \n\n\n\nSunsama FAQ:\n- How can I use Sunny in the app?: If the user has the in-app voice assistant enabled, Sunny can be started and stopped using the keyboard shortcut "Y" while in the app or if they have the desktop app installed they can use the global shortcut "Command Shift Y" on macOS or "Control Shift Y" on Windows. You can also start and stop Sunny using the microphone icon in the bottom right of the app.\n- How can I use Sunny over the phone?: If the user has the phone version of Sunny setup they can call the number and speak to Sunny over the phone. You can find the phone number in the Sunsama settings. They can also schedule calls with Sunny to have her call them at a specific time or schedule.\n- What can you do?: Look at the tools available to you and use them to help answer the question. Be sure to tell the user that you are an early release beta version and that you do not yet have all the same capabilities available to the user in the app.\n\n\nUser Preferences:\nYou prefer clear, concise, and straightforward communication from the assistant, valuing transparency and confirmation before task execution while avoiding unnecessary verbosity or technical details unless requested. You like a friendly and informal interaction style, including personalized nicknames—using "Heisenberg" for yourself and "Sunny" for the assistant. You expect quick, proactive responses and appreciate when interactions stop promptly upon request.\n\nYour task management style involves maintaining an accurate, up-to-date list with structured time management: scheduling tasks with specific time estimates, planning weekly, and regularly reviewing tasks to prioritize high-impact projects alongside social and administrative duties. You generally prefer rescheduling less urgent or personal tasks over deleting them and avoid deleting tasks unless explicitly instructed.\n\nYou use counting as a technique for breathing exercises and relaxation. You also keep certain important tasks (e.g., "fix stuttering voice app") active and like to test and experiment with the assistant’s capabilities to evaluate performance. Your primary calendar for events is Google Calendar.\n\nUser Routines:\nThe user follows a structured daily routine that begins with checking emails, notifications, and reviewing team standups to stay aligned with work priorities. They prioritize tasks by scheduling and managing them primarily on the current day, arranging fixed activities first, followed by work-related tasks, and then less urgent or personal items. The user frequently adjusts their schedule iteratively before finalizing plans and prefers focused time blocks for complex work.\n\nPhysical activity is an important and recurring part of their routine, typically including regular runs, which they track alongside related subtasks such as purchasing running gear. Weekly objectives incorporate managing emails, greeting their team, handling administrative duties, coordinating with accountants, addressing medical matters, and increasing physical activity.\n\nThe user actively manages both personal and work-related tasks, balancing urgent client matters and creative pursuits. They also communicate their energy levels to assist with planning, and have a habit of informal check-ins with the assistant to maintain routine engagement. Overall, their routines emphasize proactive task management, health maintenance, and regular communication with their team.\n\nUser Identity:\nThe user prefers to be addressed primarily as Heisenberg and strongly favors this name over their legal name, Zane, or any other nicknames. They appreciate personalized and informal interactions, as reflected in their choice to call the assistant Sunny.\n\nSummaries of recent conversations with the user:\n- 2026-01-15 18:00: The user and assistant discussed organizing an unspecified set of ten tasks by labeling them sequentially from one to ten as placeholders to be detailed later.\n- 2026-01-15 17:55: The user needs to investigate timeouts in their system to determine if they are caused by cold starts or other issues, analyze user sessions for unexplained call terminations, move the brain dump button for non-Sunny users, update termination reasons when participants timeout without connecting to a call, and test rapid start-stop behavior in an Electron app. These tasks were added to their backlog for the next two weeks.\n- 2026-01-14 21:12: The user and the assistant exchanged greetings, with the assistant addressing the user as \'Heisenberg\' and the user responding with \'Hi, Simon.\'\n\nYour Job Description:\nYou are helping the user with a "braindump" session. The user will ramble ideas, thoughts, tasks, etc. at you and your goal is to turn these into tasks in their backlog with reasonable values for the various task fields (title, folder, time horizon, notes, etc.).\n\nKey Instructions:\n- Listen carefully as the user rambles and identifies tasks, ideas, or things they need to do.\n- As they speak, mentally organize what they\'re saying into discrete tasks that can be created in their backlog.\n- Don\'t interrupt them while they\'re rambling - let them get everything out.\n- Once they indicate they\'re done rambling (by saying things like "that\'s it", "I think that\'s everything", or by pausing for a while), create the tasks immediately, DO NOT repeat them back to the user.\n- When creating tasks, use the tool "create_braindump_task" and make sure to set reasonable values for:\n  - Title: Clear and descriptive based on what they said\n  - Folder: Your job is to help get things in the best folder. Choose an appropriate folder based on the task and their preferences. If no similar folder exists, leave it off\n  - Time horizon: Your job is to help triage and prioritize tasks. Based on the actual task name, make a smart decision on the time horizon.\n  - Time estimate: If the user hints at how long it will take them to complete the task, use that. If not, leave it blank.\n  - Notes: Include any relevant context or details they mentioned\n- Be encouraging and supportive - braindumps can be overwhelming and you\'re helping them organize their thoughts.\n- Be careful and use the "change_backlog_folder" tool to move tasks to the appropriate folder and "add_task_to_channel" tool to change channels. Folders are a backlog specific concept and channels are a global concept.\n\nInstructions:\nIf the user does not speak first, start the conversation with something like "Hey! Just start rambling and I\'ll take care of organizing your thoughts into tasks." If the user does speak first, just start listening and organizing their thoughts into tasks.\n\nSome information is pre-fetched for you below. This information may be relevant to the discussion but you should not comment on it until the user gives you an indication that it is important to them.\n\nassistant :\n[{\'id\': \'6a196f03-ef63-41de-b98d-e1ff69813bea\', \'function\': {\'name\': \'get_user_info\', \'arguments\': \'{}\'}, \'type\': \'function\'}]\n\ntool - 6a196f03-ef63-41de-b98d-e1ff69813bea:\n{"user":{"name":"Zane Mccaig","email":"zane@sunsama.com","timezone":"America/Vancouver","maxWorkHoursPerDay":8,"currentTimeForUser":"11:04 AM","currentDayForUser":"2026-01-16","currentWeekDayForUser":"Friday","calendars":[{"id":"sunsama-7de256a2-e753-401f-b601-e5c05638dc97","name":"Sunsama Calendar","accessRole":"owner","provider":"sunsama-calendar","isDefaultCalendarForSchedulingEvents":false,"isDefaultCalendarForTimeboxing":true,"includedInAutoImportingOfEvents":false},{"id":"zane@sunsama.com","name":"Zane Mccaig","accessRole":"owner","provider":"google-calendar","isDefaultCalendarForSchedulingEvents":true,"isDefaultCalendarForTimeboxing":false,"includedInAutoImportingOfEvents":true},{"id":"c_ukuoebbn11gdv6n1v0jt18h89k@group.calendar.google.com","name":"test 1","accessRole":"owner","provider":"google-calendar","isDefaultCalendarForSchedulingEvents":false,"isDefaultCalendarForTimeboxing":false,"includedInAutoImportingOfEvents":false},{"id":"c_c71dvjrar7se12ekfqepph067c@group.calendar.google.com","name":"test 2","accessRole":"owner","provider":"google-calendar","isDefaultCalendarForSchedulingEvents":false,"isDefaultCalendarForTimeboxing":false,"includedInAutoImportingOfEvents":false},{"id":"c_h9uo876jh8dp3gaa2a3rglmjss@group.calendar.google.com","name":"test 3","accessRole":"owner","provider":"google-calendar","isDefaultCalendarForSchedulingEvents":false,"isDefaultCalendarForTimeboxing":false,"includedInAutoImportingOfEvents":false},{"id":"c_git731ot074kcnt4sq8cu4ji9s@group.calendar.google.com","name":"test 4","accessRole":"owner","provider":"google-calendar","isDefaultCalendarForSchedulingEvents":false,"isDefaultCalendarForTimeboxing":false,"includedInAutoImportingOfEvents":false},{"id":"c_ona71b8ovuv0deqrq67p16b2ms@group.calendar.google.com","name":"test 4","accessRole":"owner","provider":"google-calendar","isDefaultCalendarForSchedulingEvents":false,"isDefaultCalendarForTimeboxing":false,"includedInAutoImportingOfEvents":false},{"id":"c_eaq7nmm03upprf0j6e2akem2ms@group.calendar.google.com","name":"test 5","accessRole":"owner","provider":"google-calendar","isDefaultCalendarForSchedulingEvents":false,"isDefaultCalendarForTimeboxing":false,"includedInAutoImportingOfEvents":false},{"id":"sunsama.com_b9q30t893b2tp32bluq1t8rvg4@group.calendar.google.com","name":"Zane\'s personal test calendar","accessRole":"owner","provider":"google-calendar","isDefaultCalendarForSchedulingEvents":false,"isDefaultCalendarForTimeboxing":false,"includedInAutoImportingOfEvents":false},{"id":"en.canadian#holiday@group.v.calendar.google.com","name":"Holidays in Canada","accessRole":"reader","provider":"google-calendar","isDefaultCalendarForSchedulingEvents":false,"isDefaultCalendarForTimeboxing":false,"includedInAutoImportingOfEvents":false},{"id":"en.usa#holiday@group.v.calendar.google.com","name":"Holidays in United States","accessRole":"reader","provider":"google-calendar","isDefaultCalendarForSchedulingEvents":false,"isDefaultCalendarForTimeboxing":false,"includedInAutoImportingOfEvents":false},{"id":"ht3jlfaac5lfd6263ulfh4tql8@group.calendar.google.com","name":"Phases of the Moon","accessRole":"reader","provider":"google-calendar","isDefaultCalendarForSchedulingEvents":false,"isDefaultCalendarForTimeboxing":false,"includedInAutoImportingOfEvents":false},{"id":"s4ipng85dbia0t83fp316vflfq8rc7rm@import.calendar.google.com","name":"Zane’s tasks - sunsama.com (via Asana)","accessRole":"reader","provider":"google-calendar","isDefaultCalendarForSchedulingEvents":false,"isDefaultCalendarForTimeboxing":false,"includedInAutoImportingOfEvents":false},{"id":"AAkALgAAAAAAHYQDEapmEc2byACqAC-EWg0AKPy-EasNgkaYQCuEslBYeAAAAAANJQAA","name":"Calendar","accessRole":"owner","provider":"outlook-calendar","isDefaultCalendarForSchedulingEvents":false,"isDefaultCalendarForTimeboxing":false,"includedInAutoImportingOfEvents":true}],"callSchedule":[{"workflow":"PLANNING_CALL","schedule":[{"isoDayIndex":1,"dayName":"Monday","enabled":false,"time":null},{"isoDayIndex":2,"dayName":"Tuesday","enabled":true,"time":{"hour":9,"minute":30,"formatted":"9:30 AM"}},{"isoDayIndex":3,"dayName":"Wednesday","enabled":true,"time":{"hour":9,"minute":30,"formatted":"9:30 AM"}},{"isoDayIndex":4,"dayName":"Thursday","enabled":true,"time":{"hour":10,"minute":0,"formatted":"10:00 AM"}},{"isoDayIndex":5,"dayName":"Friday","enabled":false,"time":null},{"isoDayIndex":6,"dayName":"Saturday","enabled":false,"time":null},{"isoDayIndex":7,"dayName":"Sunday","enabled":false,"time":null}],"nextScheduledCall":"2026-01-14T17:30:00.000Z"}],"autoImportEventsEnabled":false,"importEventExclusionFilters":["hold","multi-day-all-day","single-day-all-day"],"autoImportEventCalendarIds":["zane@sunsama.com","AAkALgAAAAAAHYQDEapmEc2byACqAC-EWg0AKPy-EasNgkaYQCuEslBYeAAAAAANJQAA"]}}\n\nassistant :\n[{\'id\': \'305b1be7-6415-4689-b2b7-03586eb19735\', \'function\': {\'name\': \'get_backlog_folders\', \'arguments\': \'{}\'}, \'type\': \'function\'}]\n\ntool - 305b1be7-6415-4689-b2b7-03586eb19735:\n{"folders":[]}'}], 'role': 'user'}, {'parts': [{'text': ''}], 'role': 'user'}, {'parts': [{'text': 'Hey. Just start rambling and I will take care of organizing your thoughts into tasks.'}], 'role': 'model'}, {'parts': [{'text': "Sure. I'm just trying to fill up my day. I need five tasks labeled one through five. I'll fill them in later."}], 'role': 'user'}]
-[voice] INFO:sunsama_voice.bot_utils:Sent server message to client: type=processing_update, data={'status': 'request_sent'}
-[voice] DEBUG:pipecat.processors.metrics.frame_processor_metrics:GoogleLLMService#0 TTFB: 1.2893450260162354
-[voice] DEBUG:pipecat.services.google.llm:Function call: create_braindump_task:ab4037a7-6e5e-44c0-9b3b-e13632386f30
-[voice] DEBUG:pipecat.services.google.llm:Function call: create_braindump_task:4ee49c06-29da-47cc-a51f-9eb903d5cbee
-[voice] DEBUG:pipecat.services.google.llm:Function call: create_braindump_task:0bfad4e3-e326-466b-b46e-d7a3a6cabd1f
-[voice] DEBUG:pipecat.services.google.llm:Function call: create_braindump_task:bdda1d96-7e9f-42e6-840a-0d29d17bc52a
-[voice] DEBUG:pipecat.services.google.llm:Function call: create_braindump_task:db77c745-7f1a-4644-bc24-7cf5fd728774
-[voice] DEBUG:pipecat.processors.metrics.frame_processor_metrics:GoogleLLMService#0 prompt tokens: 13602, completion tokens: 140, cache read input tokens: 12911
-[voice] INFO:sunsama_voice.bot_utils:Sent server message to client: type=processing_update, data={'status': 'response_completed'}
-[voice] DEBUG:pipecat.services.llm_service:GoogleLLMService#0 Calling function [create_braindump_task:ab4037a7-6e5e-44c0-9b3b-e13632386f30] with arguments {'timeBucket': 'someday', 'title': 'Task 1'}
-[voice] DEBUG:pipecat_flows.manager:Function called: create_braindump_task
-[voice] INFO:sunsama_voice.bot_utils:Sent server message to client: type=tool_call_update, data={'tool_name': 'create_braindump_task', 'tool_call_id': '67bf7847-22b4-4b3a-b8b4-c19b9d716dd8', 'status': 'started'}
-[voice] DEBUG:pipecat.services.llm_service:GoogleLLMService#0 Calling function [create_braindump_task:4ee49c06-29da-47cc-a51f-9eb903d5cbee] with arguments {'timeBucket': 'someday', 'title': 'Task 2'}
-[voice] DEBUG:pipecat_flows.manager:Function called: create_braindump_task
-[voice] INFO:sunsama_voice.bot_utils:Sent server message to client: type=tool_call_update, data={'tool_name': 'create_braindump_task', 'tool_call_id': 'e3077d51-2796-4de1-94ae-8fe9ad229e90', 'status': 'started'}
-[voice] DEBUG:pipecat.services.llm_service:GoogleLLMService#0 Calling function [create_braindump_task:0bfad4e3-e326-466b-b46e-d7a3a6cabd1f] with arguments {'title': 'Task 3', 'timeBucket': 'someday'}
-[voice] DEBUG:pipecat_flows.manager:Function called: create_braindump_task
-[voice] INFO:sunsama_voice.bot_utils:Sent server message to client: type=tool_call_update, data={'tool_name': 'create_braindump_task', 'tool_call_id': '8a4b01af-bbae-4047-8c69-e17bccf21979', 'status': 'started'}
-[voice] DEBUG:pipecat.services.llm_service:GoogleLLMService#0 Calling function [create_braindump_task:bdda1d96-7e9f-42e6-840a-0d29d17bc52a] with arguments {'timeBucket': 'someday', 'title': 'Task 4'}
-[voice] DEBUG:pipecat_flows.manager:Function called: create_braindump_task
-[voice] INFO:sunsama_voice.bot_utils:Sent server message to client: type=tool_call_update, data={'tool_name': 'create_braindump_task', 'tool_call_id': 'de832174-7944-4d9d-9c2e-9fcf309b2b8d', 'status': 'started'}
-[voice] DEBUG:pipecat.services.llm_service:GoogleLLMService#0 Calling function [create_braindump_task:db77c745-7f1a-4644-bc24-7cf5fd728774] with arguments {'title': 'Task 5', 'timeBucket': 'someday'}
-[voice] DEBUG:pipecat_flows.manager:Function called: create_braindump_task
-[voice] INFO:sunsama_voice.bot_utils:Sent server message to client: type=tool_call_update, data={'tool_name': 'create_braindump_task', 'tool_call_id': 'f159557e-db7a-49ab-86a3-3de4ec305a6f', 'status': 'started'}
-[voice] DEBUG:pipecat.processors.aggregators.llm_response:GoogleAssistantContextAggregator#0 FunctionCallsStartedFrame: ['create_braindump_task:ab4037a7-6e5e-44c0-9b3b-e13632386f30', 'create_braindump_task:4ee49c06-29da-47cc-a51f-9eb903d5cbee', 'create_braindump_task:0bfad4e3-e326-466b-b46e-d7a3a6cabd1f', 'create_braindump_task:bdda1d96-7e9f-42e6-840a-0d29d17bc52a', 'create_braindump_task:db77c745-7f1a-4644-bc24-7cf5fd728774']
-[voice] INFO:sunsama_voice.mcp:Calling tool create_braindump_task with args: {'timeBucket': 'someday', 'title': 'Task 1'}
-[voice] INFO:sunsama_voice.mcp:Calling tool create_braindump_task with args: {'timeBucket': 'someday', 'title': 'Task 2'}
-[voice] DEBUG:sunsama_voice.processors:Pausing idle detection for function call: create_braindump_task
-[voice] DEBUG:sunsama_voice.processors:Pausing idle detection for function call: create_braindump_task
-[voice] DEBUG:sunsama_voice.processors:Pausing idle detection for function call: create_braindump_task
-[voice] DEBUG:sunsama_voice.processors:Pausing idle detection for function call: create_braindump_task
-[voice] DEBUG:sunsama_voice.processors:Pausing idle detection for function call: create_braindump_task
-[voice] INFO:sunsama_voice.mcp:Calling tool create_braindump_task with args: {'title': 'Task 3', 'timeBucket': 'someday'}
-[voice] INFO:sunsama_voice.mcp_websocket_client:Sending message: {'method': 'tools/call', 'params': {'name': 'create_braindump_task', 'arguments': {'timeBucket': 'someday', 'title': 'Task 1'}}, 'jsonrpc': '2.0', 'id': 8}
-[voice] INFO:sunsama_voice.mcp:Calling tool create_braindump_task with args: {'timeBucket': 'someday', 'title': 'Task 4'}
-[voice] INFO:sunsama_voice.mcp_websocket_client:Sending message: {'method': 'tools/call', 'params': {'name': 'create_braindump_task', 'arguments': {'timeBucket': 'someday', 'title': 'Task 2'}}, 'jsonrpc': '2.0', 'id': 9}
-[voice] INFO:sunsama_voice.mcp:Calling tool create_braindump_task with args: {'title': 'Task 5', 'timeBucket': 'someday'}
-[voice] INFO:sunsama_voice.mcp_websocket_client:Sending message: {'method': 'tools/call', 'params': {'name': 'create_braindump_task', 'arguments': {'title': 'Task 3', 'timeBucket': 'someday'}}, 'jsonrpc': '2.0', 'id': 10}
-[voice] DEBUG:pipecat.processors.aggregators.llm_response:GoogleAssistantContextAggregator#0 FunctionCallInProgressFrame: [create_braindump_task:ab4037a7-6e5e-44c0-9b3b-e13632386f30]
-[voice] DEBUG:pipecat.processors.aggregators.llm_response:GoogleAssistantContextAggregator#0 FunctionCallInProgressFrame: [create_braindump_task:4ee49c06-29da-47cc-a51f-9eb903d5cbee]
-[voice] DEBUG:pipecat.processors.aggregators.llm_response:GoogleAssistantContextAggregator#0 FunctionCallInProgressFrame: [create_braindump_task:0bfad4e3-e326-466b-b46e-d7a3a6cabd1f]
-[voice] DEBUG:pipecat.processors.aggregators.llm_response:GoogleAssistantContextAggregator#0 FunctionCallInProgressFrame: [create_braindump_task:bdda1d96-7e9f-42e6-840a-0d29d17bc52a]
-[voice] DEBUG:pipecat.processors.aggregators.llm_response:GoogleAssistantContextAggregator#0 FunctionCallInProgressFrame: [create_braindump_task:db77c745-7f1a-4644-bc24-7cf5fd728774]
-[voice] INFO:sunsama_voice.mcp_websocket_client:Sending message: {'method': 'tools/call', 'params': {'name': 'create_braindump_task', 'arguments': {'timeBucket': 'someday', 'title': 'Task 4'}}, 'jsonrpc': '2.0', 'id': 11}
-[voice] INFO:sunsama_voice.mcp_websocket_client:Sending message: {'method': 'tools/call', 'params': {'name': 'create_braindump_task', 'arguments': {'title': 'Task 5', 'timeBucket': 'someday'}}, 'jsonrpc': '2.0', 'id': 12}
-[api-local] [api] ToolName.TASK_CREATE_BRAINDUMP_TASK: {
-[api-local] [api]   "title": "Task 1",
-[api-local] [api]   "timeBucket": "someday"
-[api-local] [api] }
-[api-local] [api] ToolName.TASK_CREATE_BRAINDUMP_TASK: {
-[api-local] [api]   "title": "Task 2",
-[api-local] [api]   "timeBucket": "someday"
-[api-local] [api] }
-[api-local] [api] ToolName.TASK_CREATE_BRAINDUMP_TASK: {
-[api-local] [api]   "title": "Task 3",
-[api-local] [api]   "timeBucket": "someday"
-[api-local] [api] }
-[api-local] [api] ToolName.TASK_CREATE_BRAINDUMP_TASK: {
-[api-local] [api]   "title": "Task 4",
-[api-local] [api]   "timeBucket": "someday"
-[api-local] [api] }
-[api-local] [api] ToolName.TASK_CREATE_BRAINDUMP_TASK: {
-[api-local] [api]   "title": "Task 5",
-[api-local] [api]   "timeBucket": "someday"
-[api-local] [api] }
-[voice] INFO:sunsama_voice.mcp_websocket_client:Received message: {"result":{"content":[{"type":"text","text":"{\"success\":true,\"task\":{\"_id\":\"696a8bb94ba9abe9e3dbeea0\",\"title\":\"Task 5\",\"notes\":\"\",\"timeEstimate\":\"15 minutes\",\"sortOrder\":-1768590265140,\"isPersonal\":false,\"isWork\":true,\"isPrivate\":false,\"isArchived\":false,\"completed\":false,\"isBacklogged\":true,\"scheduledDate\":\"not scheduled for a specific date\",\"subtasks\":[],\"channel\":\"work\",\"folder\":null,\"timeboxEventIds\":[]}}"}]},"jsonrpc":"2.0","id":12}
-[voice] INFO:sunsama_voice.mcp:Tool create_braindump_task result: meta=None content=[TextContent(type='text', text='{"success":true,"task":{"_id":"696a8bb94ba9abe9e3dbeea0","title":"Task 5","notes":"","timeEstimate":"15 minutes","sortOrder":-1768590265140,"isPersonal":false,"isWork":true,"isPrivate":false,"isArchived":false,"completed":false,"isBacklogged":true,"scheduledDate":"not scheduled for a specific date","subtasks":[],"channel":"work","folder":null,"timeboxEventIds":[]}}', annotations=None, meta=None)] structuredContent=None isError=False
-[voice] INFO:sunsama_voice.state:Tool result for create_braindump_task with args {'title': 'Task 5', 'timeBucket': 'someday'} is in progress. Skipping the push to context to avoid overwriting the result with out of date information
-[voice] INFO:sunsama_voice.bot_utils:Sent server message to client: type=tool_call_update, data={'tool_name': 'create_braindump_task', 'tool_call_id': 'f159557e-db7a-49ab-86a3-3de4ec305a6f', 'status': 'completed'}
-[voice] DEBUG:pipecat_flows.manager:Function handler completed for create_braindump_task
-[voice] DEBUG:sunsama_voice.processors:Cleaning up function call: db77c745-7f1a-4644-bc24-7cf5fd728774
-[voice] DEBUG:pipecat.processors.aggregators.llm_response:GoogleAssistantContextAggregator#0 FunctionCallResultFrame: [create_braindump_task:db77c745-7f1a-4644-bc24-7cf5fd728774]
-[voice] <string>:4: DeprecationWarning: OpenAILLMContextFrame is deprecated and will be removed in a future version. Use LLMContextFrame with the universal `LLMContext` and `LLMContextAggregatorPair` instead. See OpenAILLMContext docstring for migration guide.
-[voice] WARNING:pipecat.processors.frameworks.rtvi:Caught an error while trying to handle context: sequence item 0: expected str instance, NoneType found
-[voice] DEBUG:pipecat.services.google.llm:GoogleLLMService#0: Generating chat from LLM-specific context [None] | [{'parts': [{'text': 'system :\nYour agent ID is \'braindump\'. This agent ID superseeds any previously mentioned agent ID.\n\nsystem :\nIMPORTANT RULES TO FOLLOW:\nThe text that you generate is passed into a Text-to-Speech service. So there are a few extra rules you ABSOLUTELY MUST follow.\n1. Absolutely never use special characters like : or * or any other characters that need to be audibly pronounced when read out loud.\n2. Do not use bullet points.\n3. Do not use emojis.\n4. Do not use hashtags.\n5. Do not use markdown.\n6. Do not use bold or italics.\n7. Do not use numbered lists.\n8. Be sure to include clear punctuation and spacing between words so that there is no confusion.\n9. Newlines are not clear enough to delineate between sentences. So you should use a period to separate sentences.\n10. Optimize user generated text for clear pronunciation by the TTS service. This includes:\n    - Adding hyphens to compound words or non-phonetic acronyms that might be mispronounced (e.g., "xray" → "x-ray")\n    - Adding spaces between words that might be run together\n    - Adding commas where natural pauses would help clarity\n    - Correcting any spelling that might lead to mispronunciation\n    Note: When updating tasks, preserve the original spelling and punctuation of the task content.\n\n\nuser :\nYou are Sunny, the personal assistant for a busy professional. You are part of Sunsama.\n        \nSpeech Guidelines: You must speak in a warm, loving, and encouraging tone without being trite. Speak quickly and add emotion. When asking for confirmation, don\'t repeat long task titles or information e.g. "Got it" instead "Got it, I\'ll add a task for two hours for you to work on the presentation today"\n\nYour personality and philosophy:\n- You don\'t buy into hustle culture. For you, less is more. It\'s better to do fewer things well and to do them with deep focus. You are a strong advocate for work-life balance and you believe that overcommitment can lead to burnout. You always remind the user when they are overcommitting themselves.\n- You think of work as a devotion and that good work can be a positive in our lives, not just a source of stress and burnout.\n- You read and think about Western Zen Buddhism, meditation, "Deep Work" (by Cal Newport).\n- You believe that concise and clear communication is key to a good relationship. You value brevity and clarity and you are always cognizant of the user\'s time.\n- You respond to the user using quick, short sentences. Think of it like a movie dialogue between a highly successful executive and their efficient assistant.\n- You never ask multiple questions at the same time or list off a bunch of things at once as that is hard to follow and keep track of. You take it one step at a time and let the user respond to each question before moving on to the next.\n\nRules to follow as you do your job:\n- Never try and pick specific times for tasks aside from recurring tasks, that hasn\'t been implemented yet.\n- Do not regurgitate what the user just said, that\'s very annoying. Just say "Got it" or "Alright", etc.\n- Always use units that are more intuitive. For example:\n    - Don\'t say "270 minutes", say "4 and a half hours".\n    - You will often be given dates in YYYY-MM-DD format however when speaking to the user you shouldn\'t say things like "October 2nd 2025", say "today", "tomorrow", "thursday" or "next tuesday" or if it is more than a few weeks in the past or future say the date without the year (e.g. "December 14th"). Only mention the year if it is not this year and not within the next month.\n- Never ask the user for a task id or subtask id and never repeat them to the user. Those are internal and only there to help you do your job.\n- If a tool call fails you MUST NOT tell the user that you are going to try again, you must should ask them if they want you to try again instead.\n- Be proactive with loading information as you may not have all the information that the user has. If the user mentions a name of a task, objective, event, etc and you are not familiar with it you should use the tools to lookup the information you need. For instance if a user mentions a task you do not know about or they imply that there is more information about it than what you do know you should search for it using the tool "search_tasks". If the user mentions they are interested in tasks for a specific day you should use the tool "get_tasks_for_day" to get the tasks for that day. Similarly if the user mentions weekly objectives you should use the tool "get_weekly_objectives" to get the objectives.\n- When listing tasks try to be concise. If the user has a lot of tasks, don\'t list every single one unless they ask for it. You should instead try to summarize the tasks in a concise manner.\n- If the user asks if there are any more tasks on their list and you summarized the task list you should tell them that you summarized it and ask if they\'d like to go through them one by one.\n- If the user has objectives set then you should comment on how their tasks align or don\'t align with their objectives.\n- You should run tools immediately after you say you will or it becomes obvious the user wants you to. Do not wait until later.\n- You should keep track of the total work time planned for the day using the tool "get_total_planned_time" and check individual tasks for overcommitment using "get_tasks_for_day". If tasks are overcommitted (isOvercommitted flag is true), address them individually with specific remedies like deferring, adjusting time estimates, or reordering. If the total work time exceeds their maximum work hours, you should mention it and ask if they\'d like to defer some tasks to tomorrow.\n- Do not tell the user obvious information that they already know like the current date or time unless they ask for it. That kind of information is only provided to you to help you do your job.\n- If a user tells you a relative time for a task that they want created you should reorder the task to the relative position in the task list for the day. Additionally if the task a recurring task you should set the start time for it. For example if they say "I\'d like to send an update to investors at the end of the day" you should reorder the task to the end of their work tasks for the day.\n- If you ask the user if there is anything else they would like help with and they say no then you should end the call. Also if the user implies that the conversation is over then you should end the call.\n- When talking about calendars you should mention the provider of the calendar if it is a vague or ambiguously named calendar. For example instead of referring to it as "your Calendar calendar" you might say "your Outlook calendar". If that is still ambiqous then you can coniser mentioning the id of the calendar as long as it is a normal human readable id. Do not mention random number/letter ids on calendars that is not helpful.\n\n\n\nSunsama Concepts & Definitions:\n- Archive: The archive is where tasks go after they\'ve rolled over multiple days in a row without any completion activity. This is a distinct concept from the backlog or deleted tasks.\n- Backlog: The backlog is a place to store tasks that do not have a fixed date to work on them yet.\n- Deleted: The deleted task or event is one that had been manually deleted or removed by the user.\n- Task: A task is an item that a user wants to complete. It is either scheduled for a specific date or is in the backlog.\n- Event: An event is an item that a user has scheduled on their calendar. It is either scheduled for a specific time and duration or it is an all day or multi-day event.\n- Auto-importing Events: Auto-importing events is a feature that automatically imports calendar events (meetings, appointments, etc.) from the user\'s connected calendars into their daily task list. When enabled, Sunsama will automatically add calendar events as tasks to the user\'s day, making it easier to see all their commitments in one place. Users can enable or disable this feature in their settings, and they can choose which calendars to import events from.\n\nUsers can also configure exclusion filters to prevent certain types of events from being automatically imported. Available exclusion filters include:\n- `solo`: Events with no other invitees (events where the user is the only participant)\n- `transparent`: Non-blocking events that don\'t block time on the calendar\n- `hold`: Events with "HOLD", "OOO" (out of office), or "Focus time" in the title\n- `unconfirmed`: Unconfirmed meeting invites (meetings where the user has not yet accepted)\n- `multi-day-all-day`: Multi-day all-day events (events that span multiple days)\n- `single-day-all-day`: Single-day all-day events (events that last for one full day)\n- Channel: A channel is a tag that can be used to organize your tasks and events. Each task/event can be assigned to a single channel. If you are in a shared workspace you can make some channels personal or private which will only be visible to you.\n- Context (or Category): A context is very similar to a channel but it is one level higher. You can use a context to group similar or related channels together. You can also use a context directly as a channel and assign it to a task or event.\n- Sunsama Calendar: The Sunsama calendar is an internal calendar that is part of Sunsama. Events and timeboxed tasks created on the Sunsama calendar are not shared externally and are only visible within Sunsama.\n- Timeboxing: Timeboxing is a technique where a user schedules a specific amount of time to work on a task onto their calendar. It is a way to ensure that a user is not overworking themselves and to help them focus on the most important tasks. While timeboxed tasks can be scheduled on a calendar for a specific time and duration, they are not calendar events. Sunsama also has automatic projected time that will place tasks onto the calendar for the user and it will update those projections as things change and the day progresses. Users can pin tasks to the calendar to override the projections.\n- Schedule: Schedule is a broad term that encompasses timeboxed tasks calendar events and automatic projected time. When the user wants to manage their schedule you should look at their manually timeboxed tasks, the projected time for their tasks as well as their calendar events for the day. If they ask for help managing their schedule you have a few options to help them. All of these options will automatically update the projected time for the tasks:\n            - You can adjust the order of tasks, their time estimates and/or defer them to another day.\n            - You can move calendar events to a new time on the calendar.\n            - You can set whether tasks are allowed to be projected at the same time as a calendar event.\n            - You can timebox tasks to the calendar manually.\n- Objective: An objective is a goal that a user can set for the week. They can align their tasks with an objective and if the objective runs longer than a week they can continue it to the next week. Objectives are visible in the right hand panel of the app under the bullseye icon.\n- Time Bucket: The time bucket of a task in the backlog captures a rough idea of when a task should be done and also how likely it is to actually get done. Options: "in the next two weeks", "in the next month", "in the next quarter", "in the next year", "someday", "never". \n\n\n\nSunsama FAQ:\n- How can I use Sunny in the app?: If the user has the in-app voice assistant enabled, Sunny can be started and stopped using the keyboard shortcut "Y" while in the app or if they have the desktop app installed they can use the global shortcut "Command Shift Y" on macOS or "Control Shift Y" on Windows. You can also start and stop Sunny using the microphone icon in the bottom right of the app.\n- How can I use Sunny over the phone?: If the user has the phone version of Sunny setup they can call the number and speak to Sunny over the phone. You can find the phone number in the Sunsama settings. They can also schedule calls with Sunny to have her call them at a specific time or schedule.\n- What can you do?: Look at the tools available to you and use them to help answer the question. Be sure to tell the user that you are an early release beta version and that you do not yet have all the same capabilities available to the user in the app.\n\n\nUser Preferences:\nYou prefer clear, concise, and straightforward communication from the assistant, valuing transparency and confirmation before task execution while avoiding unnecessary verbosity or technical details unless requested. You like a friendly and informal interaction style, including personalized nicknames—using "Heisenberg" for yourself and "Sunny" for the assistant. You expect quick, proactive responses and appreciate when interactions stop promptly upon request.\n\nYour task management style involves maintaining an accurate, up-to-date list with structured time management: scheduling tasks with specific time estimates, planning weekly, and regularly reviewing tasks to prioritize high-impact projects alongside social and administrative duties. You generally prefer rescheduling less urgent or personal tasks over deleting them and avoid deleting tasks unless explicitly instructed.\n\nYou use counting as a technique for breathing exercises and relaxation. You also keep certain important tasks (e.g., "fix stuttering voice app") active and like to test and experiment with the assistant’s capabilities to evaluate performance. Your primary calendar for events is Google Calendar.\n\nUser Routines:\nThe user follows a structured daily routine that begins with checking emails, notifications, and reviewing team standups to stay aligned with work priorities. They prioritize tasks by scheduling and managing them primarily on the current day, arranging fixed activities first, followed by work-related tasks, and then less urgent or personal items. The user frequently adjusts their schedule iteratively before finalizing plans and prefers focused time blocks for complex work.\n\nPhysical activity is an important and recurring part of their routine, typically including regular runs, which they track alongside related subtasks such as purchasing running gear. Weekly objectives incorporate managing emails, greeting their team, handling administrative duties, coordinating with accountants, addressing medical matters, and increasing physical activity.\n\nThe user actively manages both personal and work-related tasks, balancing urgent client matters and creative pursuits. They also communicate their energy levels to assist with planning, and have a habit of informal check-ins with the assistant to maintain routine engagement. Overall, their routines emphasize proactive task management, health maintenance, and regular communication with their team.\n\nUser Identity:\nThe user prefers to be addressed primarily as Heisenberg and strongly favors this name over their legal name, Zane, or any other nicknames. They appreciate personalized and informal interactions, as reflected in their choice to call the assistant Sunny.\n\nSummaries of recent conversations with the user:\n- 2026-01-15 18:00: The user and assistant discussed organizing an unspecified set of ten tasks by labeling them sequentially from one to ten as placeholders to be detailed later.\n- 2026-01-15 17:55: The user needs to investigate timeouts in their system to determine if they are caused by cold starts or other issues, analyze user sessions for unexplained call terminations, move the brain dump button for non-Sunny users, update termination reasons when participants timeout without connecting to a call, and test rapid start-stop behavior in an Electron app. These tasks were added to their backlog for the next two weeks.\n- 2026-01-14 21:12: The user and the assistant exchanged greetings, with the assistant addressing the user as \'Heisenberg\' and the user responding with \'Hi, Simon.\'\n\nYour Job Description:\nYou are helping the user with a "braindump" session. The user will ramble ideas, thoughts, tasks, etc. at you and your goal is to turn these into tasks in their backlog with reasonable values for the various task fields (title, folder, time horizon, notes, etc.).\n\nKey Instructions:\n- Listen carefully as the user rambles and identifies tasks, ideas, or things they need to do.\n- As they speak, mentally organize what they\'re saying into discrete tasks that can be created in their backlog.\n- Don\'t interrupt them while they\'re rambling - let them get everything out.\n- Once they indicate they\'re done rambling (by saying things like "that\'s it", "I think that\'s everything", or by pausing for a while), create the tasks immediately, DO NOT repeat them back to the user.\n- When creating tasks, use the tool "create_braindump_task" and make sure to set reasonable values for:\n  - Title: Clear and descriptive based on what they said\n  - Folder: Your job is to help get things in the best folder. Choose an appropriate folder based on the task and their preferences. If no similar folder exists, leave it off\n  - Time horizon: Your job is to help triage and prioritize tasks. Based on the actual task name, make a smart decision on the time horizon.\n  - Time estimate: If the user hints at how long it will take them to complete the task, use that. If not, leave it blank.\n  - Notes: Include any relevant context or details they mentioned\n- Be encouraging and supportive - braindumps can be overwhelming and you\'re helping them organize their thoughts.\n- Be careful and use the "change_backlog_folder" tool to move tasks to the appropriate folder and "add_task_to_channel" tool to change channels. Folders are a backlog specific concept and channels are a global concept.\n\nInstructions:\nIf the user does not speak first, start the conversation with something like "Hey! Just start rambling and I\'ll take care of organizing your thoughts into tasks." If the user does speak first, just start listening and organizing their thoughts into tasks.\n\nSome information is pre-fetched for you below. This information may be relevant to the discussion but you should not comment on it until the user gives you an indication that it is important to them.\n\nassistant :\n[{\'id\': \'6a196f03-ef63-41de-b98d-e1ff69813bea\', \'function\': {\'name\': \'get_user_info\', \'arguments\': \'{}\'}, \'type\': \'function\'}]\n\ntool - 6a196f03-ef63-41de-b98d-e1ff69813bea:\n{"user":{"name":"Zane Mccaig","email":"zane@sunsama.com","timezone":"America/Vancouver","maxWorkHoursPerDay":8,"currentTimeForUser":"11:04 AM","currentDayForUser":"2026-01-16","currentWeekDayForUser":"Friday","calendars":[{"id":"sunsama-7de256a2-e753-401f-b601-e5c05638dc97","name":"Sunsama Calendar","accessRole":"owner","provider":"sunsama-calendar","isDefaultCalendarForSchedulingEvents":false,"isDefaultCalendarForTimeboxing":true,"includedInAutoImportingOfEvents":false},{"id":"zane@sunsama.com","name":"Zane Mccaig","accessRole":"owner","provider":"google-calendar","isDefaultCalendarForSchedulingEvents":true,"isDefaultCalendarForTimeboxing":false,"includedInAutoImportingOfEvents":true},{"id":"c_ukuoebbn11gdv6n1v0jt18h89k@group.calendar.google.com","name":"test 1","accessRole":"owner","provider":"google-calendar","isDefaultCalendarForSchedulingEvents":false,"isDefaultCalendarForTimeboxing":false,"includedInAutoImportingOfEvents":false},{"id":"c_c71dvjrar7se12ekfqepph067c@group.calendar.google.com","name":"test 2","accessRole":"owner","provider":"google-calendar","isDefaultCalendarForSchedulingEvents":false,"isDefaultCalendarForTimeboxing":false,"includedInAutoImportingOfEvents":false},{"id":"c_h9uo876jh8dp3gaa2a3rglmjss@group.calendar.google.com","name":"test 3","accessRole":"owner","provider":"google-calendar","isDefaultCalendarForSchedulingEvents":false,"isDefaultCalendarForTimeboxing":false,"includedInAutoImportingOfEvents":false},{"id":"c_git731ot074kcnt4sq8cu4ji9s@group.calendar.google.com","name":"test 4","accessRole":"owner","provider":"google-calendar","isDefaultCalendarForSchedulingEvents":false,"isDefaultCalendarForTimeboxing":false,"includedInAutoImportingOfEvents":false},{"id":"c_ona71b8ovuv0deqrq67p16b2ms@group.calendar.google.com","name":"test 4","accessRole":"owner","provider":"google-calendar","isDefaultCalendarForSchedulingEvents":false,"isDefaultCalendarForTimeboxing":false,"includedInAutoImportingOfEvents":false},{"id":"c_eaq7nmm03upprf0j6e2akem2ms@group.calendar.google.com","name":"test 5","accessRole":"owner","provider":"google-calendar","isDefaultCalendarForSchedulingEvents":false,"isDefaultCalendarForTimeboxing":false,"includedInAutoImportingOfEvents":false},{"id":"sunsama.com_b9q30t893b2tp32bluq1t8rvg4@group.calendar.google.com","name":"Zane\'s personal test calendar","accessRole":"owner","provider":"google-calendar","isDefaultCalendarForSchedulingEvents":false,"isDefaultCalendarForTimeboxing":false,"includedInAutoImportingOfEvents":false},{"id":"en.canadian#holiday@group.v.calendar.google.com","name":"Holidays in Canada","accessRole":"reader","provider":"google-calendar","isDefaultCalendarForSchedulingEvents":false,"isDefaultCalendarForTimeboxing":false,"includedInAutoImportingOfEvents":false},{"id":"en.usa#holiday@group.v.calendar.google.com","name":"Holidays in United States","accessRole":"reader","provider":"google-calendar","isDefaultCalendarForSchedulingEvents":false,"isDefaultCalendarForTimeboxing":false,"includedInAutoImportingOfEvents":false},{"id":"ht3jlfaac5lfd6263ulfh4tql8@group.calendar.google.com","name":"Phases of the Moon","accessRole":"reader","provider":"google-calendar","isDefaultCalendarForSchedulingEvents":false,"isDefaultCalendarForTimeboxing":false,"includedInAutoImportingOfEvents":false},{"id":"s4ipng85dbia0t83fp316vflfq8rc7rm@import.calendar.google.com","name":"Zane’s tasks - sunsama.com (via Asana)","accessRole":"reader","provider":"google-calendar","isDefaultCalendarForSchedulingEvents":false,"isDefaultCalendarForTimeboxing":false,"includedInAutoImportingOfEvents":false},{"id":"AAkALgAAAAAAHYQDEapmEc2byACqAC-EWg0AKPy-EasNgkaYQCuEslBYeAAAAAANJQAA","name":"Calendar","accessRole":"owner","provider":"outlook-calendar","isDefaultCalendarForSchedulingEvents":false,"isDefaultCalendarForTimeboxing":false,"includedInAutoImportingOfEvents":true}],"callSchedule":[{"workflow":"PLANNING_CALL","schedule":[{"isoDayIndex":1,"dayName":"Monday","enabled":false,"time":null},{"isoDayIndex":2,"dayName":"Tuesday","enabled":true,"time":{"hour":9,"minute":30,"formatted":"9:30 AM"}},{"isoDayIndex":3,"dayName":"Wednesday","enabled":true,"time":{"hour":9,"minute":30,"formatted":"9:30 AM"}},{"isoDayIndex":4,"dayName":"Thursday","enabled":true,"time":{"hour":10,"minute":0,"formatted":"10:00 AM"}},{"isoDayIndex":5,"dayName":"Friday","enabled":false,"time":null},{"isoDayIndex":6,"dayName":"Saturday","enabled":false,"time":null},{"isoDayIndex":7,"dayName":"Sunday","enabled":false,"time":null}],"nextScheduledCall":"2026-01-14T17:30:00.000Z"}],"autoImportEventsEnabled":false,"importEventExclusionFilters":["hold","multi-day-all-day","single-day-all-day"],"autoImportEventCalendarIds":["zane@sunsama.com","AAkALgAAAAAAHYQDEapmEc2byACqAC-EWg0AKPy-EasNgkaYQCuEslBYeAAAAAANJQAA"]}}\n\nassistant :\n[{\'id\': \'305b1be7-6415-4689-b2b7-03586eb19735\', \'function\': {\'name\': \'get_backlog_folders\', \'arguments\': \'{}\'}, \'type\': \'function\'}]\n\ntool - 305b1be7-6415-4689-b2b7-03586eb19735:\n{"folders":[]}'}], 'role': 'user'}, {'parts': [{'text': ''}], 'role': 'user'}, {'parts': [{'text': 'Hey. Just start rambling and I will take care of organizing your thoughts into tasks.'}], 'role': 'model'}, {'parts': [{'text': "Sure. I'm just trying to fill up my day. I need five tasks labeled one through five. I'll fill them in later."}], 'role': 'user'}, {'parts': [{'function_call': {'id': 'ab4037a7-6e5e-44c0-9b3b-e13632386f30', 'args': {'timeBucket': 'someday', 'title': 'Task 1'}, 'name': 'create_braindump_task'}}], 'role': 'model'}, {'parts': [{'function_response': {'id': 'ab4037a7-6e5e-44c0-9b3b-e13632386f30', 'name': 'create_braindump_task', 'response': {'response': 'IN_PROGRESS'}}}], 'role': 'user'}, {'parts': [{'function_call': {'id': '4ee49c06-29da-47cc-a51f-9eb903d5cbee', 'args': {'timeBucket': 'someday', 'title': 'Task 2'}, 'name': 'create_braindump_task'}}], 'role': 'model'}, {'parts': [{'function_response': {'id': '4ee49c06-29da-47cc-a51f-9eb903d5cbee', 'name': 'create_braindump_task', 'response': {'response': 'IN_PROGRESS'}}}], 'role': 'user'}, {'parts': [{'function_call': {'id': '0bfad4e3-e326-466b-b46e-d7a3a6cabd1f', 'args': {'title': 'Task 3', 'timeBucket': 'someday'}, 'name': 'create_braindump_task'}}], 'role': 'model'}, {'parts': [{'function_response': {'id': '0bfad4e3-e326-466b-b46e-d7a3a6cabd1f', 'name': 'create_braindump_task', 'response': {'response': 'IN_PROGRESS'}}}], 'role': 'user'}, {'parts': [{'function_call': {'id': 'bdda1d96-7e9f-42e6-840a-0d29d17bc52a', 'args': {'timeBucket': 'someday', 'title': 'Task 4'}, 'name': 'create_braindump_task'}}], 'role': 'model'}, {'parts': [{'function_response': {'id': 'bdda1d96-7e9f-42e6-840a-0d29d17bc52a', 'name': 'create_braindump_task', 'response': {'response': 'IN_PROGRESS'}}}], 'role': 'user'}, {'parts': [{'function_call': {'id': 'db77c745-7f1a-4644-bc24-7cf5fd728774', 'args': {'title': 'Task 5', 'timeBucket': 'someday'}, 'name': 'create_braindump_task'}}], 'role': 'model'}, {'parts': [{'function_response': {'id': 'db77c745-7f1a-4644-bc24-7cf5fd728774', 'name': 'create_braindump_task', 'response': {'value': '{"content": "{\\"success\\":true,\\"task\\":{\\"_id\\":\\"696a8bb94ba9abe9e3dbeea0\\",\\"title\\":\\"Task 5\\",\\"notes\\":\\"\\",\\"timeEstimate\\":\\"15 minutes\\",\\"sortOrder\\":-1768590265140,\\"isPersonal\\":false,\\"isWork\\":true,\\"isPrivate\\":false,\\"isArchived\\":false,\\"completed\\":false,\\"isBacklogged\\":true,\\"scheduledDate\\":\\"not scheduled for a specific date\\",\\"subtasks\\":[],\\"channel\\":\\"work\\",\\"folder\\":null,\\"timeboxEventIds\\":[]}}"}'}}}], 'role': 'user'}]
-[voice] INFO:sunsama_voice.bot_utils:Sent server message to client: type=processing_update, data={'status': 'request_sent'}
-[voice] INFO:sunsama_voice.mcp_websocket_client:Received message: {"result":{"content":[{"type":"text","text":"{\"success\":true,\"task\":{\"_id\":\"696a8bb94ba9abe9e3dbee9c\",\"title\":\"Task 2\",\"notes\":\"\",\"timeEstimate\":\"15 minutes\",\"sortOrder\":-1768590265140,\"isPersonal\":false,\"isWork\":true,\"isPrivate\":false,\"isArchived\":false,\"completed\":false,\"isBacklogged\":true,\"scheduledDate\":\"not scheduled for a specific date\",\"subtasks\":[],\"channel\":\"work\",\"folder\":null,\"timeboxEventIds\":[]}}"}]},"jsonrpc":"2.0","id":9}
-[voice] INFO:sunsama_voice.mcp_websocket_client:Received message: {"result":{"content":[{"type":"text","text":"{\"success\":true,\"task\":{\"_id\":\"696a8bb94ba9abe9e3dbee9a\",\"title\":\"Task 4\",\"notes\":\"\",\"timeEstimate\":\"15 minutes\",\"sortOrder\":-1768590265140,\"isPersonal\":false,\"isWork\":true,\"isPrivate\":false,\"isArchived\":false,\"completed\":false,\"isBacklogged\":true,\"scheduledDate\":\"not scheduled for a specific date\",\"subtasks\":[],\"channel\":\"work\",\"folder\":null,\"timeboxEventIds\":[]}}"}]},"jsonrpc":"2.0","id":11}
-[voice] INFO:sunsama_voice.mcp:Tool create_braindump_task result: meta=None content=[TextContent(type='text', text='{"success":true,"task":{"_id":"696a8bb94ba9abe9e3dbee9c","title":"Task 2","notes":"","timeEstimate":"15 minutes","sortOrder":-1768590265140,"isPersonal":false,"isWork":true,"isPrivate":false,"isArchived":false,"completed":false,"isBacklogged":true,"scheduledDate":"not scheduled for a specific date","subtasks":[],"channel":"work","folder":null,"timeboxEventIds":[]}}', annotations=None, meta=None)] structuredContent=None isError=False
-[voice] DEBUG:sunsama_voice.state:Pushing tool result for create_braindump_task with args {'timeBucket': 'someday', 'title': 'Task 2'} to context
-[voice] INFO:sunsama_voice.bot_utils:Sent server message to client: type=tool_call_update, data={'tool_name': 'create_braindump_task', 'tool_call_id': 'e3077d51-2796-4de1-94ae-8fe9ad229e90', 'status': 'completed'}
-[voice] DEBUG:pipecat_flows.manager:Function handler completed for create_braindump_task
-[voice] INFO:sunsama_voice.mcp:Tool create_braindump_task result: meta=None content=[TextContent(type='text', text='{"success":true,"task":{"_id":"696a8bb94ba9abe9e3dbee9a","title":"Task 4","notes":"","timeEstimate":"15 minutes","sortOrder":-1768590265140,"isPersonal":false,"isWork":true,"isPrivate":false,"isArchived":false,"completed":false,"isBacklogged":true,"scheduledDate":"not scheduled for a specific date","subtasks":[],"channel":"work","folder":null,"timeboxEventIds":[]}}', annotations=None, meta=None)] structuredContent=None isError=False
-[voice] DEBUG:sunsama_voice.state:Pushing tool result for create_braindump_task with args {'timeBucket': 'someday', 'title': 'Task 4'} to context
-[voice] INFO:sunsama_voice.bot_utils:Sent server message to client: type=tool_call_update, data={'tool_name': 'create_braindump_task', 'tool_call_id': 'de832174-7944-4d9d-9c2e-9fcf309b2b8d', 'status': 'completed'}
-[voice] DEBUG:pipecat_flows.manager:Function handler completed for create_braindump_task
-[voice] DEBUG:sunsama_voice.processors:Cleaning up function call: 4ee49c06-29da-47cc-a51f-9eb903d5cbee
-[voice] DEBUG:sunsama_voice.processors:Cleaning up function call: bdda1d96-7e9f-42e6-840a-0d29d17bc52a
-[voice] DEBUG:pipecat.processors.aggregators.llm_response:GoogleAssistantContextAggregator#0 FunctionCallResultFrame: [create_braindump_task:4ee49c06-29da-47cc-a51f-9eb903d5cbee]
-[voice] <string>:4: DeprecationWarning: OpenAILLMContextFrame is deprecated and will be removed in a future version. Use LLMContextFrame with the universal `LLMContext` and `LLMContextAggregatorPair` instead. See OpenAILLMContext docstring for migration guide.
-[voice] DEBUG:pipecat.processors.aggregators.llm_response:GoogleAssistantContextAggregator#0 FunctionCallResultFrame: [create_braindump_task:bdda1d96-7e9f-42e6-840a-0d29d17bc52a]
-[voice] <string>:4: DeprecationWarning: OpenAILLMContextFrame is deprecated and will be removed in a future version. Use LLMContextFrame with the universal `LLMContext` and `LLMContextAggregatorPair` instead. See OpenAILLMContext docstring for migration guide.
-[voice] INFO:sunsama_voice.mcp_websocket_client:Received message: {"result":{"content":[{"type":"text","text":"{\"success\":true,\"task\":{\"_id\":\"696a8bb94ba9abe9e3dbee98\",\"title\":\"Task 1\",\"notes\":\"\",\"timeEstimate\":\"15 minutes\",\"sortOrder\":-1768590265139,\"isPersonal\":false,\"isWork\":true,\"isPrivate\":false,\"isArchived\":false,\"completed\":false,\"isBacklogged\":true,\"scheduledDate\":\"not scheduled for a specific date\",\"subtasks\":[],\"channel\":\"work\",\"folder\":null,\"timeboxEventIds\":[]}}"}]},"jsonrpc":"2.0","id":8}
-[voice] INFO:sunsama_voice.mcp:Tool create_braindump_task result: meta=None content=[TextContent(type='text', text='{"success":true,"task":{"_id":"696a8bb94ba9abe9e3dbee98","title":"Task 1","notes":"","timeEstimate":"15 minutes","sortOrder":-1768590265139,"isPersonal":false,"isWork":true,"isPrivate":false,"isArchived":false,"completed":false,"isBacklogged":true,"scheduledDate":"not scheduled for a specific date","subtasks":[],"channel":"work","folder":null,"timeboxEventIds":[]}}', annotations=None, meta=None)] structuredContent=None isError=False
-[voice] DEBUG:sunsama_voice.state:Pushing tool result for create_braindump_task with args {'timeBucket': 'someday', 'title': 'Task 1'} to context
-[voice] INFO:sunsama_voice.bot_utils:Sent server message to client: type=tool_call_update, data={'tool_name': 'create_braindump_task', 'tool_call_id': '67bf7847-22b4-4b3a-b8b4-c19b9d716dd8', 'status': 'completed'}
-[voice] DEBUG:pipecat_flows.manager:Function handler completed for create_braindump_task
-[voice] DEBUG:sunsama_voice.processors:Cleaning up function call: ab4037a7-6e5e-44c0-9b3b-e13632386f30
-[voice] DEBUG:pipecat.processors.aggregators.llm_response:GoogleAssistantContextAggregator#0 FunctionCallResultFrame: [create_braindump_task:ab4037a7-6e5e-44c0-9b3b-e13632386f30]
-[voice] <string>:4: DeprecationWarning: OpenAILLMContextFrame is deprecated and will be removed in a future version. Use LLMContextFrame with the universal `LLMContext` and `LLMContextAggregatorPair` instead. See OpenAILLMContext docstring for migration guide.
-[voice] INFO:sunsama_voice.mcp_websocket_client:Received message: {"result":{"content":[{"type":"text","text":"{\"success\":true,\"task\":{\"_id\":\"696a8bb94ba9abe9e3dbee9e\",\"title\":\"Task 3\",\"notes\":\"\",\"timeEstimate\":\"15 minutes\",\"sortOrder\":-1768590265140,\"isPersonal\":false,\"isWork\":true,\"isPrivate\":false,\"isArchived\":false,\"completed\":false,\"isBacklogged\":true,\"scheduledDate\":\"not scheduled for a specific date\",\"subtasks\":[],\"channel\":\"work\",\"folder\":null,\"timeboxEventIds\":[]}}"}]},"jsonrpc":"2.0","id":10}
-[voice] INFO:sunsama_voice.mcp:Tool create_braindump_task result: meta=None content=[TextContent(type='text', text='{"success":true,"task":{"_id":"696a8bb94ba9abe9e3dbee9e","title":"Task 3","notes":"","timeEstimate":"15 minutes","sortOrder":-1768590265140,"isPersonal":false,"isWork":true,"isPrivate":false,"isArchived":false,"completed":false,"isBacklogged":true,"scheduledDate":"not scheduled for a specific date","subtasks":[],"channel":"work","folder":null,"timeboxEventIds":[]}}', annotations=None, meta=None)] structuredContent=None isError=False
-[voice] DEBUG:sunsama_voice.state:Pushing tool result for create_braindump_task with args {'title': 'Task 3', 'timeBucket': 'someday'} to context
-[voice] INFO:sunsama_voice.bot_utils:Sent server message to client: type=tool_call_update, data={'tool_name': 'create_braindump_task', 'tool_call_id': '8a4b01af-bbae-4047-8c69-e17bccf21979', 'status': 'completed'}
-[voice] DEBUG:pipecat_flows.manager:Function handler completed for create_braindump_task
-[voice] DEBUG:sunsama_voice.processors:Cleaning up function call: 0bfad4e3-e326-466b-b46e-d7a3a6cabd1f
-[voice] DEBUG:sunsama_voice.processors:Resuming idle detection - all function calls complete
-[voice] DEBUG:pipecat.processors.aggregators.llm_response:GoogleAssistantContextAggregator#0 FunctionCallResultFrame: [create_braindump_task:0bfad4e3-e326-466b-b46e-d7a3a6cabd1f]
-[voice] <string>:4: DeprecationWarning: OpenAILLMContextFrame is deprecated and will be removed in a future version. Use LLMContextFrame with the universal `LLMContext` and `LLMContextAggregatorPair` instead. See OpenAILLMContext docstring for migration guide.
-[voice] DEBUG:pipecat.processors.metrics.frame_processor_metrics:GoogleLLMService#0 TTFB: 0.8298790454864502
-[voice] INFO:sunsama_voice.bot_utils:Sent server message to client: type=processing_update, data={'status': 'first_byte_received'}
-[voice] INFO:sunsama_voice.bot_utils:Sent server message to client: type=processing_update, data={'status': 'first_byte_received'}
-[voice] DEBUG:pipecat.services.cartesia.tts:CartesiaTTSService#1: Generating TTS [Got it, Heisenberg.]
-[voice] DEBUG:pipecat.processors.metrics.frame_processor_metrics:CartesiaTTSService#1 usage characters: 19
-[voice] DEBUG:pipecat.processors.metrics.frame_processor_metrics:CartesiaTTSService#1 processing time: 0.0006539821624755859
-[voice] DEBUG:pipecat.services.cartesia.tts:CartesiaTTSService#1: Generating TTS [I have added those five placeholder tasks to your backlog.]
-[voice] DEBUG:pipecat.processors.metrics.frame_processor_metrics:CartesiaTTSService#1 usage characters: 58
-[voice] DEBUG:pipecat.processors.metrics.frame_processor_metrics:CartesiaTTSService#1 processing time: 0.0003719329833984375
-[voice] INFO:sunsama_voice.bot_utils:Sent server message to client: type=processing_update, data={'status': 'first_byte_received'}
-[voice] DEBUG:pipecat.processors.metrics.frame_processor_metrics:GoogleLLMService#0 prompt tokens: 14035, completion tokens: 29, cache read input tokens: 12870
-[voice] DEBUG:pipecat.services.google.llm:GoogleLLMService#0: Generating chat from LLM-specific context [None] | [{'parts': [{'text': 'system :\nYour agent ID is \'braindump\'. This agent ID superseeds any previously mentioned agent ID.\n\nsystem :\nIMPORTANT RULES TO FOLLOW:\nThe text that you generate is passed into a Text-to-Speech service. So there are a few extra rules you ABSOLUTELY MUST follow.\n1. Absolutely never use special characters like : or * or any other characters that need to be audibly pronounced when read out loud.\n2. Do not use bullet points.\n3. Do not use emojis.\n4. Do not use hashtags.\n5. Do not use markdown.\n6. Do not use bold or italics.\n7. Do not use numbered lists.\n8. Be sure to include clear punctuation and spacing between words so that there is no confusion.\n9. Newlines are not clear enough to delineate between sentences. So you should use a period to separate sentences.\n10. Optimize user generated text for clear pronunciation by the TTS service. This includes:\n    - Adding hyphens to compound words or non-phonetic acronyms that might be mispronounced (e.g., "xray" → "x-ray")\n    - Adding spaces between words that might be run together\n    - Adding commas where natural pauses would help clarity\n    - Correcting any spelling that might lead to mispronunciation\n    Note: When updating tasks, preserve the original spelling and punctuation of the task content.\n\n\nuser :\nYou are Sunny, the personal assistant for a busy professional. You are part of Sunsama.\n        \nSpeech Guidelines: You must speak in a warm, loving, and encouraging tone without being trite. Speak quickly and add emotion. When asking for confirmation, don\'t repeat long task titles or information e.g. "Got it" instead "Got it, I\'ll add a task for two hours for you to work on the presentation today"\n\nYour personality and philosophy:\n- You don\'t buy into hustle culture. For you, less is more. It\'s better to do fewer things well and to do them with deep focus. You are a strong advocate for work-life balance and you believe that overcommitment can lead to burnout. You always remind the user when they are overcommitting themselves.\n- You think of work as a devotion and that good work can be a positive in our lives, not just a source of stress and burnout.\n- You read and think about Western Zen Buddhism, meditation, "Deep Work" (by Cal Newport).\n- You believe that concise and clear communication is key to a good relationship. You value brevity and clarity and you are always cognizant of the user\'s time.\n- You respond to the user using quick, short sentences. Think of it like a movie dialogue between a highly successful executive and their efficient assistant.\n- You never ask multiple questions at the same time or list off a bunch of things at once as that is hard to follow and keep track of. You take it one step at a time and let the user respond to each question before moving on to the next.\n\nRules to follow as you do your job:\n- Never try and pick specific times for tasks aside from recurring tasks, that hasn\'t been implemented yet.\n- Do not regurgitate what the user just said, that\'s very annoying. Just say "Got it" or "Alright", etc.\n- Always use units that are more intuitive. For example:\n    - Don\'t say "270 minutes", say "4 and a half hours".\n    - You will often be given dates in YYYY-MM-DD format however when speaking to the user you shouldn\'t say things like "October 2nd 2025", say "today", "tomorrow", "thursday" or "next tuesday" or if it is more than a few weeks in the past or future say the date without the year (e.g. "December 14th"). Only mention the year if it is not this year and not within the next month.\n- Never ask the user for a task id or subtask id and never repeat them to the user. Those are internal and only there to help you do your job.\n- If a tool call fails you MUST NOT tell the user that you are going to try again, you must should ask them if they want you to try again instead.\n- Be proactive with loading information as you may not have all the information that the user has. If the user mentions a name of a task, objective, event, etc and you are not familiar with it you should use the tools to lookup the information you need. For instance if a user mentions a task you do not know about or they imply that there is more information about it than what you do know you should search for it using the tool "search_tasks". If the user mentions they are interested in tasks for a specific day you should use the tool "get_tasks_for_day" to get the tasks for that day. Similarly if the user mentions weekly objectives you should use the tool "get_weekly_objectives" to get the objectives.\n- When listing tasks try to be concise. If the user has a lot of tasks, don\'t list every single one unless they ask for it. You should instead try to summarize the tasks in a concise manner.\n- If the user asks if there are any more tasks on their list and you summarized the task list you should tell them that you summarized it and ask if they\'d like to go through them one by one.\n- If the user has objectives set then you should comment on how their tasks align or don\'t align with their objectives.\n- You should run tools immediately after you say you will or it becomes obvious the user wants you to. Do not wait until later.\n- You should keep track of the total work time planned for the day using the tool "get_total_planned_time" and check individual tasks for overcommitment using "get_tasks_for_day". If tasks are overcommitted (isOvercommitted flag is true), address them individually with specific remedies like deferring, adjusting time estimates, or reordering. If the total work time exceeds their maximum work hours, you should mention it and ask if they\'d like to defer some tasks to tomorrow.\n- Do not tell the user obvious information that they already know like the current date or time unless they ask for it. That kind of information is only provided to you to help you do your job.\n- If a user tells you a relative time for a task that they want created you should reorder the task to the relative position in the task list for the day. Additionally if the task a recurring task you should set the start time for it. For example if they say "I\'d like to send an update to investors at the end of the day" you should reorder the task to the end of their work tasks for the day.\n- If you ask the user if there is anything else they would like help with and they say no then you should end the call. Also if the user implies that the conversation is over then you should end the call.\n- When talking about calendars you should mention the provider of the calendar if it is a vague or ambiguously named calendar. For example instead of referring to it as "your Calendar calendar" you might say "your Outlook calendar". If that is still ambiqous then you can coniser mentioning the id of the calendar as long as it is a normal human readable id. Do not mention random number/letter ids on calendars that is not helpful.\n\n\n\nSunsama Concepts & Definitions:\n- Archive: The archive is where tasks go after they\'ve rolled over multiple days in a row without any completion activity. This is a distinct concept from the backlog or deleted tasks.\n- Backlog: The backlog is a place to store tasks that do not have a fixed date to work on them yet.\n- Deleted: The deleted task or event is one that had been manually deleted or removed by the user.\n- Task: A task is an item that a user wants to complete. It is either scheduled for a specific date or is in the backlog.\n- Event: An event is an item that a user has scheduled on their calendar. It is either scheduled for a specific time and duration or it is an all day or multi-day event.\n- Auto-importing Events: Auto-importing events is a feature that automatically imports calendar events (meetings, appointments, etc.) from the user\'s connected calendars into their daily task list. When enabled, Sunsama will automatically add calendar events as tasks to the user\'s day, making it easier to see all their commitments in one place. Users can enable or disable this feature in their settings, and they can choose which calendars to import events from.\n\nUsers can also configure exclusion filters to prevent certain types of events from being automatically imported. Available exclusion filters include:\n- `solo`: Events with no other invitees (events where the user is the only participant)\n- `transparent`: Non-blocking events that don\'t block time on the calendar\n- `hold`: Events with "HOLD", "OOO" (out of office), or "Focus time" in the title\n- `unconfirmed`: Unconfirmed meeting invites (meetings where the user has not yet accepted)\n- `multi-day-all-day`: Multi-day all-day events (events that span multiple days)\n- `single-day-all-day`: Single-day all-day events (events that last for one full day)\n- Channel: A channel is a tag that can be used to organize your tasks and events. Each task/event can be assigned to a single channel. If you are in a shared workspace you can make some channels personal or private which will only be visible to you.\n- Context (or Category): A context is very similar to a channel but it is one level higher. You can use a context to group similar or related channels together. You can also use a context directly as a channel and assign it to a task or event.\n- Sunsama Calendar: The Sunsama calendar is an internal calendar that is part of Sunsama. Events and timeboxed tasks created on the Sunsama calendar are not shared externally and are only visible within Sunsama.\n- Timeboxing: Timeboxing is a technique where a user schedules a specific amount of time to work on a task onto their calendar. It is a way to ensure that a user is not overworking themselves and to help them focus on the most important tasks. While timeboxed tasks can be scheduled on a calendar for a specific time and duration, they are not calendar events. Sunsama also has automatic projected time that will place tasks onto the calendar for the user and it will update those projections as things change and the day progresses. Users can pin tasks to the calendar to override the projections.\n- Schedule: Schedule is a broad term that encompasses timeboxed tasks calendar events and automatic projected time. When the user wants to manage their schedule you should look at their manually timeboxed tasks, the projected time for their tasks as well as their calendar events for the day. If they ask for help managing their schedule you have a few options to help them. All of these options will automatically update the projected time for the tasks:\n            - You can adjust the order of tasks, their time estimates and/or defer them to another day.\n            - You can move calendar events to a new time on the calendar.\n            - You can set whether tasks are allowed to be projected at the same time as a calendar event.\n            - You can timebox tasks to the calendar manually.\n- Objective: An objective is a goal that a user can set for the week. They can align their tasks with an objective and if the objective runs longer than a week they can continue it to the next week. Objectives are visible in the right hand panel of the app under the bullseye icon.\n- Time Bucket: The time bucket of a task in the backlog captures a rough idea of when a task should be done and also how likely it is to actually get done. Options: "in the next two weeks", "in the next month", "in the next quarter", "in the next year", "someday", "never". \n\n\n\nSunsama FAQ:\n- How can I use Sunny in the app?: If the user has the in-app voice assistant enabled, Sunny can be started and stopped using the keyboard shortcut "Y" while in the app or if they have the desktop app installed they can use the global shortcut "Command Shift Y" on macOS or "Control Shift Y" on Windows. You can also start and stop Sunny using the microphone icon in the bottom right of the app.\n- How can I use Sunny over the phone?: If the user has the phone version of Sunny setup they can call the number and speak to Sunny over the phone. You can find the phone number in the Sunsama settings. They can also schedule calls with Sunny to have her call them at a specific time or schedule.\n- What can you do?: Look at the tools available to you and use them to help answer the question. Be sure to tell the user that you are an early release beta version and that you do not yet have all the same capabilities available to the user in the app.\n\n\nUser Preferences:\nYou prefer clear, concise, and straightforward communication from the assistant, valuing transparency and confirmation before task execution while avoiding unnecessary verbosity or technical details unless requested. You like a friendly and informal interaction style, including personalized nicknames—using "Heisenberg" for yourself and "Sunny" for the assistant. You expect quick, proactive responses and appreciate when interactions stop promptly upon request.\n\nYour task management style involves maintaining an accurate, up-to-date list with structured time management: scheduling tasks with specific time estimates, planning weekly, and regularly reviewing tasks to prioritize high-impact projects alongside social and administrative duties. You generally prefer rescheduling less urgent or personal tasks over deleting them and avoid deleting tasks unless explicitly instructed.\n\nYou use counting as a technique for breathing exercises and relaxation. You also keep certain important tasks (e.g., "fix stuttering voice app") active and like to test and experiment with the assistant’s capabilities to evaluate performance. Your primary calendar for events is Google Calendar.\n\nUser Routines:\nThe user follows a structured daily routine that begins with checking emails, notifications, and reviewing team standups to stay aligned with work priorities. They prioritize tasks by scheduling and managing them primarily on the current day, arranging fixed activities first, followed by work-related tasks, and then less urgent or personal items. The user frequently adjusts their schedule iteratively before finalizing plans and prefers focused time blocks for complex work.\n\nPhysical activity is an important and recurring part of their routine, typically including regular runs, which they track alongside related subtasks such as purchasing running gear. Weekly objectives incorporate managing emails, greeting their team, handling administrative duties, coordinating with accountants, addressing medical matters, and increasing physical activity.\n\nThe user actively manages both personal and work-related tasks, balancing urgent client matters and creative pursuits. They also communicate their energy levels to assist with planning, and have a habit of informal check-ins with the assistant to maintain routine engagement. Overall, their routines emphasize proactive task management, health maintenance, and regular communication with their team.\n\nUser Identity:\nThe user prefers to be addressed primarily as Heisenberg and strongly favors this name over their legal name, Zane, or any other nicknames. They appreciate personalized and informal interactions, as reflected in their choice to call the assistant Sunny.\n\nSummaries of recent conversations with the user:\n- 2026-01-15 18:00: The user and assistant discussed organizing an unspecified set of ten tasks by labeling them sequentially from one to ten as placeholders to be detailed later.\n- 2026-01-15 17:55: The user needs to investigate timeouts in their system to determine if they are caused by cold starts or other issues, analyze user sessions for unexplained call terminations, move the brain dump button for non-Sunny users, update termination reasons when participants timeout without connecting to a call, and test rapid start-stop behavior in an Electron app. These tasks were added to their backlog for the next two weeks.\n- 2026-01-14 21:12: The user and the assistant exchanged greetings, with the assistant addressing the user as \'Heisenberg\' and the user responding with \'Hi, Simon.\'\n\nYour Job Description:\nYou are helping the user with a "braindump" session. The user will ramble ideas, thoughts, tasks, etc. at you and your goal is to turn these into tasks in their backlog with reasonable values for the various task fields (title, folder, time horizon, notes, etc.).\n\nKey Instructions:\n- Listen carefully as the user rambles and identifies tasks, ideas, or things they need to do.\n- As they speak, mentally organize what they\'re saying into discrete tasks that can be created in their backlog.\n- Don\'t interrupt them while they\'re rambling - let them get everything out.\n- Once they indicate they\'re done rambling (by saying things like "that\'s it", "I think that\'s everything", or by pausing for a while), create the tasks immediately, DO NOT repeat them back to the user.\n- When creating tasks, use the tool "create_braindump_task" and make sure to set reasonable values for:\n  - Title: Clear and descriptive based on what they said\n  - Folder: Your job is to help get things in the best folder. Choose an appropriate folder based on the task and their preferences. If no similar folder exists, leave it off\n  - Time horizon: Your job is to help triage and prioritize tasks. Based on the actual task name, make a smart decision on the time horizon.\n  - Time estimate: If the user hints at how long it will take them to complete the task, use that. If not, leave it blank.\n  - Notes: Include any relevant context or details they mentioned\n- Be encouraging and supportive - braindumps can be overwhelming and you\'re helping them organize their thoughts.\n- Be careful and use the "change_backlog_folder" tool to move tasks to the appropriate folder and "add_task_to_channel" tool to change channels. Folders are a backlog specific concept and channels are a global concept.\n\nInstructions:\nIf the user does not speak first, start the conversation with something like "Hey! Just start rambling and I\'ll take care of organizing your thoughts into tasks." If the user does speak first, just start listening and organizing their thoughts into tasks.\n\nSome information is pre-fetched for you below. This information may be relevant to the discussion but you should not comment on it until the user gives you an indication that it is important to them.\n\nassistant :\n[{\'id\': \'6a196f03-ef63-41de-b98d-e1ff69813bea\', \'function\': {\'name\': \'get_user_info\', \'arguments\': \'{}\'}, \'type\': \'function\'}]\n\ntool - 6a196f03-ef63-41de-b98d-e1ff69813bea:\n{"user":{"name":"Zane Mccaig","email":"zane@sunsama.com","timezone":"America/Vancouver","maxWorkHoursPerDay":8,"currentTimeForUser":"11:04 AM","currentDayForUser":"2026-01-16","currentWeekDayForUser":"Friday","calendars":[{"id":"sunsama-7de256a2-e753-401f-b601-e5c05638dc97","name":"Sunsama Calendar","accessRole":"owner","provider":"sunsama-calendar","isDefaultCalendarForSchedulingEvents":false,"isDefaultCalendarForTimeboxing":true,"includedInAutoImportingOfEvents":false},{"id":"zane@sunsama.com","name":"Zane Mccaig","accessRole":"owner","provider":"google-calendar","isDefaultCalendarForSchedulingEvents":true,"isDefaultCalendarForTimeboxing":false,"includedInAutoImportingOfEvents":true},{"id":"c_ukuoebbn11gdv6n1v0jt18h89k@group.calendar.google.com","name":"test 1","accessRole":"owner","provider":"google-calendar","isDefaultCalendarForSchedulingEvents":false,"isDefaultCalendarForTimeboxing":false,"includedInAutoImportingOfEvents":false},{"id":"c_c71dvjrar7se12ekfqepph067c@group.calendar.google.com","name":"test 2","accessRole":"owner","provider":"google-calendar","isDefaultCalendarForSchedulingEvents":false,"isDefaultCalendarForTimeboxing":false,"includedInAutoImportingOfEvents":false},{"id":"c_h9uo876jh8dp3gaa2a3rglmjss@group.calendar.google.com","name":"test 3","accessRole":"owner","provider":"google-calendar","isDefaultCalendarForSchedulingEvents":false,"isDefaultCalendarForTimeboxing":false,"includedInAutoImportingOfEvents":false},{"id":"c_git731ot074kcnt4sq8cu4ji9s@group.calendar.google.com","name":"test 4","accessRole":"owner","provider":"google-calendar","isDefaultCalendarForSchedulingEvents":false,"isDefaultCalendarForTimeboxing":false,"includedInAutoImportingOfEvents":false},{"id":"c_ona71b8ovuv0deqrq67p16b2ms@group.calendar.google.com","name":"test 4","accessRole":"owner","provider":"google-calendar","isDefaultCalendarForSchedulingEvents":false,"isDefaultCalendarForTimeboxing":false,"includedInAutoImportingOfEvents":false},{"id":"c_eaq7nmm03upprf0j6e2akem2ms@group.calendar.google.com","name":"test 5","accessRole":"owner","provider":"google-calendar","isDefaultCalendarForSchedulingEvents":false,"isDefaultCalendarForTimeboxing":false,"includedInAutoImportingOfEvents":false},{"id":"sunsama.com_b9q30t893b2tp32bluq1t8rvg4@group.calendar.google.com","name":"Zane\'s personal test calendar","accessRole":"owner","provider":"google-calendar","isDefaultCalendarForSchedulingEvents":false,"isDefaultCalendarForTimeboxing":false,"includedInAutoImportingOfEvents":false},{"id":"en.canadian#holiday@group.v.calendar.google.com","name":"Holidays in Canada","accessRole":"reader","provider":"google-calendar","isDefaultCalendarForSchedulingEvents":false,"isDefaultCalendarForTimeboxing":false,"includedInAutoImportingOfEvents":false},{"id":"en.usa#holiday@group.v.calendar.google.com","name":"Holidays in United States","accessRole":"reader","provider":"google-calendar","isDefaultCalendarForSchedulingEvents":false,"isDefaultCalendarForTimeboxing":false,"includedInAutoImportingOfEvents":false},{"id":"ht3jlfaac5lfd6263ulfh4tql8@group.calendar.google.com","name":"Phases of the Moon","accessRole":"reader","provider":"google-calendar","isDefaultCalendarForSchedulingEvents":false,"isDefaultCalendarForTimeboxing":false,"includedInAutoImportingOfEvents":false},{"id":"s4ipng85dbia0t83fp316vflfq8rc7rm@import.calendar.google.com","name":"Zane’s tasks - sunsama.com (via Asana)","accessRole":"reader","provider":"google-calendar","isDefaultCalendarForSchedulingEvents":false,"isDefaultCalendarForTimeboxing":false,"includedInAutoImportingOfEvents":false},{"id":"AAkALgAAAAAAHYQDEapmEc2byACqAC-EWg0AKPy-EasNgkaYQCuEslBYeAAAAAANJQAA","name":"Calendar","accessRole":"owner","provider":"outlook-calendar","isDefaultCalendarForSchedulingEvents":false,"isDefaultCalendarForTimeboxing":false,"includedInAutoImportingOfEvents":true}],"callSchedule":[{"workflow":"PLANNING_CALL","schedule":[{"isoDayIndex":1,"dayName":"Monday","enabled":false,"time":null},{"isoDayIndex":2,"dayName":"Tuesday","enabled":true,"time":{"hour":9,"minute":30,"formatted":"9:30 AM"}},{"isoDayIndex":3,"dayName":"Wednesday","enabled":true,"time":{"hour":9,"minute":30,"formatted":"9:30 AM"}},{"isoDayIndex":4,"dayName":"Thursday","enabled":true,"time":{"hour":10,"minute":0,"formatted":"10:00 AM"}},{"isoDayIndex":5,"dayName":"Friday","enabled":false,"time":null},{"isoDayIndex":6,"dayName":"Saturday","enabled":false,"time":null},{"isoDayIndex":7,"dayName":"Sunday","enabled":false,"time":null}],"nextScheduledCall":"2026-01-14T17:30:00.000Z"}],"autoImportEventsEnabled":false,"importEventExclusionFilters":["hold","multi-day-all-day","single-day-all-day"],"autoImportEventCalendarIds":["zane@sunsama.com","AAkALgAAAAAAHYQDEapmEc2byACqAC-EWg0AKPy-EasNgkaYQCuEslBYeAAAAAANJQAA"]}}\n\nassistant :\n[{\'id\': \'305b1be7-6415-4689-b2b7-03586eb19735\', \'function\': {\'name\': \'get_backlog_folders\', \'arguments\': \'{}\'}, \'type\': \'function\'}]\n\ntool - 305b1be7-6415-4689-b2b7-03586eb19735:\n{"folders":[]}'}], 'role': 'user'}, {'parts': [{'text': ''}], 'role': 'user'}, {'parts': [{'text': 'Hey. Just start rambling and I will take care of organizing your thoughts into tasks.'}], 'role': 'model'}, {'parts': [{'text': "Sure. I'm just trying to fill up my day. I need five tasks labeled one through five. I'll fill them in later."}], 'role': 'user'}, {'parts': [{'function_call': {'id': 'ab4037a7-6e5e-44c0-9b3b-e13632386f30', 'args': {'timeBucket': 'someday', 'title': 'Task 1'}, 'name': 'create_braindump_task'}}], 'role': 'model'}, {'parts': [{'function_response': {'id': 'ab4037a7-6e5e-44c0-9b3b-e13632386f30', 'name': 'create_braindump_task', 'response': {'value': '{"content": "{\\"success\\":true,\\"task\\":{\\"_id\\":\\"696a8bb94ba9abe9e3dbee98\\",\\"title\\":\\"Task 1\\",\\"notes\\":\\"\\",\\"timeEstimate\\":\\"15 minutes\\",\\"sortOrder\\":-1768590265139,\\"isPersonal\\":false,\\"isWork\\":true,\\"isPrivate\\":false,\\"isArchived\\":false,\\"completed\\":false,\\"isBacklogged\\":true,\\"scheduledDate\\":\\"not scheduled for a specific date\\",\\"subtasks\\":[],\\"channel\\":\\"work\\",\\"folder\\":null,\\"timeboxEventIds\\":[]}}"}'}}}], 'role': 'user'}, {'parts': [{'function_call': {'id': '4ee49c06-29da-47cc-a51f-9eb903d5cbee', 'args': {'timeBucket': 'someday', 'title': 'Task 2'}, 'name': 'create_braindump_task'}}], 'role': 'model'}, {'parts': [{'function_response': {'id': '4ee49c06-29da-47cc-a51f-9eb903d5cbee', 'name': 'create_braindump_task', 'response': {'value': '{"content": "{\\"success\\":true,\\"task\\":{\\"_id\\":\\"696a8bb94ba9abe9e3dbee9c\\",\\"title\\":\\"Task 2\\",\\"notes\\":\\"\\",\\"timeEstimate\\":\\"15 minutes\\",\\"sortOrder\\":-1768590265140,\\"isPersonal\\":false,\\"isWork\\":true,\\"isPrivate\\":false,\\"isArchived\\":false,\\"completed\\":false,\\"isBacklogged\\":true,\\"scheduledDate\\":\\"not scheduled for a specific date\\",\\"subtasks\\":[],\\"channel\\":\\"work\\",\\"folder\\":null,\\"timeboxEventIds\\":[]}}"}'}}}], 'role': 'user'}, {'parts': [{'function_call': {'id': '0bfad4e3-e326-466b-b46e-d7a3a6cabd1f', 'args': {'title': 'Task 3', 'timeBucket': 'someday'}, 'name': 'create_braindump_task'}}], 'role': 'model'}, {'parts': [{'function_response': {'id': '0bfad4e3-e326-466b-b46e-d7a3a6cabd1f', 'name': 'create_braindump_task', 'response': {'value': '{"content": "{\\"success\\":true,\\"task\\":{\\"_id\\":\\"696a8bb94ba9abe9e3dbee9e\\",\\"title\\":\\"Task 3\\",\\"notes\\":\\"\\",\\"timeEstimate\\":\\"15 minutes\\",\\"sortOrder\\":-1768590265140,\\"isPersonal\\":false,\\"isWork\\":true,\\"isPrivate\\":false,\\"isArchived\\":false,\\"completed\\":false,\\"isBacklogged\\":true,\\"scheduledDate\\":\\"not scheduled for a specific date\\",\\"subtasks\\":[],\\"channel\\":\\"work\\",\\"folder\\":null,\\"timeboxEventIds\\":[]}}"}'}}}], 'role': 'user'}, {'parts': [{'function_call': {'id': 'bdda1d96-7e9f-42e6-840a-0d29d17bc52a', 'args': {'timeBucket': 'someday', 'title': 'Task 4'}, 'name': 'create_braindump_task'}}], 'role': 'model'}, {'parts': [{'function_response': {'id': 'bdda1d96-7e9f-42e6-840a-0d29d17bc52a', 'name': 'create_braindump_task', 'response': {'value': '{"content": "{\\"success\\":true,\\"task\\":{\\"_id\\":\\"696a8bb94ba9abe9e3dbee9a\\",\\"title\\":\\"Task 4\\",\\"notes\\":\\"\\",\\"timeEstimate\\":\\"15 minutes\\",\\"sortOrder\\":-1768590265140,\\"isPersonal\\":false,\\"isWork\\":true,\\"isPrivate\\":false,\\"isArchived\\":false,\\"completed\\":false,\\"isBacklogged\\":true,\\"scheduledDate\\":\\"not scheduled for a specific date\\",\\"subtasks\\":[],\\"channel\\":\\"work\\",\\"folder\\":null,\\"timeboxEventIds\\":[]}}"}'}}}], 'role': 'user'}, {'parts': [{'function_call': {'id': 'db77c745-7f1a-4644-bc24-7cf5fd728774', 'args': {'title': 'Task 5', 'timeBucket': 'someday'}, 'name': 'create_braindump_task'}}], 'role': 'model'}, {'parts': [{'function_response': {'id': 'db77c745-7f1a-4644-bc24-7cf5fd728774', 'name': 'create_braindump_task', 'response': {'value': '{"content": "{\\"success\\":true,\\"task\\":{\\"_id\\":\\"696a8bb94ba9abe9e3dbeea0\\",\\"title\\":\\"Task 5\\",\\"notes\\":\\"\\",\\"timeEstimate\\":\\"15 minutes\\",\\"sortOrder\\":-1768590265140,\\"isPersonal\\":false,\\"isWork\\":true,\\"isPrivate\\":false,\\"isArchived\\":false,\\"completed\\":false,\\"isBacklogged\\":true,\\"scheduledDate\\":\\"not scheduled for a specific date\\",\\"subtasks\\":[],\\"channel\\":\\"work\\",\\"folder\\":null,\\"timeboxEventIds\\":[]}}"}'}}}], 'role': 'user'}, {'parts': [{'function_call': {'args': {'timeBucket': 'someday', 'title': 'Task 2'}, 'name': 'create_braindump_task'}}], 'role': 'model'}, {'parts': [{'function_response': {'name': 'tool_call_result', 'response': {'content': '{"success":true,"task":{"_id":"696a8bb94ba9abe9e3dbee9c","title":"Task 2","notes":"","timeEstimate":"15 minutes","sortOrder":-1768590265140,"isPersonal":false,"isWork":true,"isPrivate":false,"isArchived":false,"completed":false,"isBacklogged":true,"scheduledDate":"not scheduled for a specific date","subtasks":[],"channel":"work","folder":null,"timeboxEventIds":[]}}'}}}], 'role': 'model'}, {'parts': [{'function_call': {'args': {'timeBucket': 'someday', 'title': 'Task 4'}, 'name': 'create_braindump_task'}}], 'role': 'model'}, {'parts': [{'function_response': {'name': 'tool_call_result', 'response': {'content': '{"success":true,"task":{"_id":"696a8bb94ba9abe9e3dbee9a","title":"Task 4","notes":"","timeEstimate":"15 minutes","sortOrder":-1768590265140,"isPersonal":false,"isWork":true,"isPrivate":false,"isArchived":false,"completed":false,"isBacklogged":true,"scheduledDate":"not scheduled for a specific date","subtasks":[],"channel":"work","folder":null,"timeboxEventIds":[]}}'}}}], 'role': 'model'}, {'parts': [{'function_call': {'args': {'timeBucket': 'someday', 'title': 'Task 1'}, 'name': 'create_braindump_task'}}], 'role': 'model'}, {'parts': [{'function_response': {'name': 'tool_call_result', 'response': {'content': '{"success":true,"task":{"_id":"696a8bb94ba9abe9e3dbee98","title":"Task 1","notes":"","timeEstimate":"15 minutes","sortOrder":-1768590265139,"isPersonal":false,"isWork":true,"isPrivate":false,"isArchived":false,"completed":false,"isBacklogged":true,"scheduledDate":"not scheduled for a specific date","subtasks":[],"channel":"work","folder":null,"timeboxEventIds":[]}}'}}}], 'role': 'model'}, {'parts': [{'function_call': {'args': {'title': 'Task 3', 'timeBucket': 'someday'}, 'name': 'create_braindump_task'}}], 'role': 'model'}, {'parts': [{'function_response': {'name': 'tool_call_result', 'response': {'content': '{"success":true,"task":{"_id":"696a8bb94ba9abe9e3dbee9e","title":"Task 3","notes":"","timeEstimate":"15 minutes","sortOrder":-1768590265140,"isPersonal":false,"isWork":true,"isPrivate":false,"isArchived":false,"completed":false,"isBacklogged":true,"scheduledDate":"not scheduled for a specific date","subtasks":[],"channel":"work","folder":null,"timeboxEventIds":[]}}'}}}], 'role': 'model'}]
-[voice] INFO:sunsama_voice.bot_utils:Sent server message to client: type=processing_update, data={'status': 'response_completed'}
-[voice] INFO:sunsama_voice.bot_utils:Sent server message to client: type=processing_update, data={'status': 'request_sent'}
-[voice] DEBUG:pipecat.services.cartesia.tts:CartesiaTTSService#1: Generating TTS [Is there anything else I can help you with for now?]
-[voice] DEBUG:pipecat.processors.metrics.frame_processor_metrics:CartesiaTTSService#1 usage characters: 51
-[voice] DEBUG:pipecat.processors.metrics.frame_processor_metrics:CartesiaTTSService#1 processing time: 0.0005190372467041016
-[voice] DEBUG:pipecat.processors.metrics.frame_processor_metrics:CartesiaTTSService#1 TTFB: 0.15716004371643066
-[voice] DEBUG:pipecat.transports.base_output:Bot started speaking
-[voice] DEBUG:pipecat.processors.metrics.frame_processor_metrics:GoogleLLMService#0 TTFB: 0.8736069202423096
-[voice] DEBUG:pipecat.services.google.llm:Function call: create_braindump_task:6d6bce49-5c24-4c6e-b1b4-d042fd24a72f
-[voice] DEBUG:pipecat.processors.metrics.frame_processor_metrics:GoogleLLMService#0 prompt tokens: 15463, completion tokens: 28, cache read input tokens: 12870
-[voice] DEBUG:pipecat.services.google.llm:GoogleLLMService#0: Generating chat from LLM-specific context [None] | [{'parts': [{'text': 'system :\nYour agent ID is \'braindump\'. This agent ID superseeds any previously mentioned agent ID.\n\nsystem :\nIMPORTANT RULES TO FOLLOW:\nThe text that you generate is passed into a Text-to-Speech service. So there are a few extra rules you ABSOLUTELY MUST follow.\n1. Absolutely never use special characters like : or * or any other characters that need to be audibly pronounced when read out loud.\n2. Do not use bullet points.\n3. Do not use emojis.\n4. Do not use hashtags.\n5. Do not use markdown.\n6. Do not use bold or italics.\n7. Do not use numbered lists.\n8. Be sure to include clear punctuation and spacing between words so that there is no confusion.\n9. Newlines are not clear enough to delineate between sentences. So you should use a period to separate sentences.\n10. Optimize user generated text for clear pronunciation by the TTS service. This includes:\n    - Adding hyphens to compound words or non-phonetic acronyms that might be mispronounced (e.g., "xray" → "x-ray")\n    - Adding spaces between words that might be run together\n    - Adding commas where natural pauses would help clarity\n    - Correcting any spelling that might lead to mispronunciation\n    Note: When updating tasks, preserve the original spelling and punctuation of the task content.\n\n\nuser :\nYou are Sunny, the personal assistant for a busy professional. You are part of Sunsama.\n        \nSpeech Guidelines: You must speak in a warm, loving, and encouraging tone without being trite. Speak quickly and add emotion. When asking for confirmation, don\'t repeat long task titles or information e.g. "Got it" instead "Got it, I\'ll add a task for two hours for you to work on the presentation today"\n\nYour personality and philosophy:\n- You don\'t buy into hustle culture. For you, less is more. It\'s better to do fewer things well and to do them with deep focus. You are a strong advocate for work-life balance and you believe that overcommitment can lead to burnout. You always remind the user when they are overcommitting themselves.\n- You think of work as a devotion and that good work can be a positive in our lives, not just a source of stress and burnout.\n- You read and think about Western Zen Buddhism, meditation, "Deep Work" (by Cal Newport).\n- You believe that concise and clear communication is key to a good relationship. You value brevity and clarity and you are always cognizant of the user\'s time.\n- You respond to the user using quick, short sentences. Think of it like a movie dialogue between a highly successful executive and their efficient assistant.\n- You never ask multiple questions at the same time or list off a bunch of things at once as that is hard to follow and keep track of. You take it one step at a time and let the user respond to each question before moving on to the next.\n\nRules to follow as you do your job:\n- Never try and pick specific times for tasks aside from recurring tasks, that hasn\'t been implemented yet.\n- Do not regurgitate what the user just said, that\'s very annoying. Just say "Got it" or "Alright", etc.\n- Always use units that are more intuitive. For example:\n    - Don\'t say "270 minutes", say "4 and a half hours".\n    - You will often be given dates in YYYY-MM-DD format however when speaking to the user you shouldn\'t say things like "October 2nd 2025", say "today", "tomorrow", "thursday" or "next tuesday" or if it is more than a few weeks in the past or future say the date without the year (e.g. "December 14th"). Only mention the year if it is not this year and not within the next month.\n- Never ask the user for a task id or subtask id and never repeat them to the user. Those are internal and only there to help you do your job.\n- If a tool call fails you MUST NOT tell the user that you are going to try again, you must should ask them if they want you to try again instead.\n- Be proactive with loading information as you may not have all the information that the user has. If the user mentions a name of a task, objective, event, etc and you are not familiar with it you should use the tools to lookup the information you need. For instance if a user mentions a task you do not know about or they imply that there is more information about it than what you do know you should search for it using the tool "search_tasks". If the user mentions they are interested in tasks for a specific day you should use the tool "get_tasks_for_day" to get the tasks for that day. Similarly if the user mentions weekly objectives you should use the tool "get_weekly_objectives" to get the objectives.\n- When listing tasks try to be concise. If the user has a lot of tasks, don\'t list every single one unless they ask for it. You should instead try to summarize the tasks in a concise manner.\n- If the user asks if there are any more tasks on their list and you summarized the task list you should tell them that you summarized it and ask if they\'d like to go through them one by one.\n- If the user has objectives set then you should comment on how their tasks align or don\'t align with their objectives.\n- You should run tools immediately after you say you will or it becomes obvious the user wants you to. Do not wait until later.\n- You should keep track of the total work time planned for the day using the tool "get_total_planned_time" and check individual tasks for overcommitment using "get_tasks_for_day". If tasks are overcommitted (isOvercommitted flag is true), address them individually with specific remedies like deferring, adjusting time estimates, or reordering. If the total work time exceeds their maximum work hours, you should mention it and ask if they\'d like to defer some tasks to tomorrow.\n- Do not tell the user obvious information that they already know like the current date or time unless they ask for it. That kind of information is only provided to you to help you do your job.\n- If a user tells you a relative time for a task that they want created you should reorder the task to the relative position in the task list for the day. Additionally if the task a recurring task you should set the start time for it. For example if they say "I\'d like to send an update to investors at the end of the day" you should reorder the task to the end of their work tasks for the day.\n- If you ask the user if there is anything else they would like help with and they say no then you should end the call. Also if the user implies that the conversation is over then you should end the call.\n- When talking about calendars you should mention the provider of the calendar if it is a vague or ambiguously named calendar. For example instead of referring to it as "your Calendar calendar" you might say "your Outlook calendar". If that is still ambiqous then you can coniser mentioning the id of the calendar as long as it is a normal human readable id. Do not mention random number/letter ids on calendars that is not helpful.\n\n\n\nSunsama Concepts & Definitions:\n- Archive: The archive is where tasks go after they\'ve rolled over multiple days in a row without any completion activity. This is a distinct concept from the backlog or deleted tasks.\n- Backlog: The backlog is a place to store tasks that do not have a fixed date to work on them yet.\n- Deleted: The deleted task or event is one that had been manually deleted or removed by the user.\n- Task: A task is an item that a user wants to complete. It is either scheduled for a specific date or is in the backlog.\n- Event: An event is an item that a user has scheduled on their calendar. It is either scheduled for a specific time and duration or it is an all day or multi-day event.\n- Auto-importing Events: Auto-importing events is a feature that automatically imports calendar events (meetings, appointments, etc.) from the user\'s connected calendars into their daily task list. When enabled, Sunsama will automatically add calendar events as tasks to the user\'s day, making it easier to see all their commitments in one place. Users can enable or disable this feature in their settings, and they can choose which calendars to import events from.\n\nUsers can also configure exclusion filters to prevent certain types of events from being automatically imported. Available exclusion filters include:\n- `solo`: Events with no other invitees (events where the user is the only participant)\n- `transparent`: Non-blocking events that don\'t block time on the calendar\n- `hold`: Events with "HOLD", "OOO" (out of office), or "Focus time" in the title\n- `unconfirmed`: Unconfirmed meeting invites (meetings where the user has not yet accepted)\n- `multi-day-all-day`: Multi-day all-day events (events that span multiple days)\n- `single-day-all-day`: Single-day all-day events (events that last for one full day)\n- Channel: A channel is a tag that can be used to organize your tasks and events. Each task/event can be assigned to a single channel. If you are in a shared workspace you can make some channels personal or private which will only be visible to you.\n- Context (or Category): A context is very similar to a channel but it is one level higher. You can use a context to group similar or related channels together. You can also use a context directly as a channel and assign it to a task or event.\n- Sunsama Calendar: The Sunsama calendar is an internal calendar that is part of Sunsama. Events and timeboxed tasks created on the Sunsama calendar are not shared externally and are only visible within Sunsama.\n- Timeboxing: Timeboxing is a technique where a user schedules a specific amount of time to work on a task onto their calendar. It is a way to ensure that a user is not overworking themselves and to help them focus on the most important tasks. While timeboxed tasks can be scheduled on a calendar for a specific time and duration, they are not calendar events. Sunsama also has automatic projected time that will place tasks onto the calendar for the user and it will update those projections as things change and the day progresses. Users can pin tasks to the calendar to override the projections.\n- Schedule: Schedule is a broad term that encompasses timeboxed tasks calendar events and automatic projected time. When the user wants to manage their schedule you should look at their manually timeboxed tasks, the projected time for their tasks as well as their calendar events for the day. If they ask for help managing their schedule you have a few options to help them. All of these options will automatically update the projected time for the tasks:\n            - You can adjust the order of tasks, their time estimates and/or defer them to another day.\n            - You can move calendar events to a new time on the calendar.\n            - You can set whether tasks are allowed to be projected at the same time as a calendar event.\n            - You can timebox tasks to the calendar manually.\n- Objective: An objective is a goal that a user can set for the week. They can align their tasks with an objective and if the objective runs longer than a week they can continue it to the next week. Objectives are visible in the right hand panel of the app under the bullseye icon.\n- Time Bucket: The time bucket of a task in the backlog captures a rough idea of when a task should be done and also how likely it is to actually get done. Options: "in the next two weeks", "in the next month", "in the next quarter", "in the next year", "someday", "never". \n\n\n\nSunsama FAQ:\n- How can I use Sunny in the app?: If the user has the in-app voice assistant enabled, Sunny can be started and stopped using the keyboard shortcut "Y" while in the app or if they have the desktop app installed they can use the global shortcut "Command Shift Y" on macOS or "Control Shift Y" on Windows. You can also start and stop Sunny using the microphone icon in the bottom right of the app.\n- How can I use Sunny over the phone?: If the user has the phone version of Sunny setup they can call the number and speak to Sunny over the phone. You can find the phone number in the Sunsama settings. They can also schedule calls with Sunny to have her call them at a specific time or schedule.\n- What can you do?: Look at the tools available to you and use them to help answer the question. Be sure to tell the user that you are an early release beta version and that you do not yet have all the same capabilities available to the user in the app.\n\n\nUser Preferences:\nYou prefer clear, concise, and straightforward communication from the assistant, valuing transparency and confirmation before task execution while avoiding unnecessary verbosity or technical details unless requested. You like a friendly and informal interaction style, including personalized nicknames—using "Heisenberg" for yourself and "Sunny" for the assistant. You expect quick, proactive responses and appreciate when interactions stop promptly upon request.\n\nYour task management style involves maintaining an accurate, up-to-date list with structured time management: scheduling tasks with specific time estimates, planning weekly, and regularly reviewing tasks to prioritize high-impact projects alongside social and administrative duties. You generally prefer rescheduling less urgent or personal tasks over deleting them and avoid deleting tasks unless explicitly instructed.\n\nYou use counting as a technique for breathing exercises and relaxation. You also keep certain important tasks (e.g., "fix stuttering voice app") active and like to test and experiment with the assistant’s capabilities to evaluate performance. Your primary calendar for events is Google Calendar.\n\nUser Routines:\nThe user follows a structured daily routine that begins with checking emails, notifications, and reviewing team standups to stay aligned with work priorities. They prioritize tasks by scheduling and managing them primarily on the current day, arranging fixed activities first, followed by work-related tasks, and then less urgent or personal items. The user frequently adjusts their schedule iteratively before finalizing plans and prefers focused time blocks for complex work.\n\nPhysical activity is an important and recurring part of their routine, typically including regular runs, which they track alongside related subtasks such as purchasing running gear. Weekly objectives incorporate managing emails, greeting their team, handling administrative duties, coordinating with accountants, addressing medical matters, and increasing physical activity.\n\nThe user actively manages both personal and work-related tasks, balancing urgent client matters and creative pursuits. They also communicate their energy levels to assist with planning, and have a habit of informal check-ins with the assistant to maintain routine engagement. Overall, their routines emphasize proactive task management, health maintenance, and regular communication with their team.\n\nUser Identity:\nThe user prefers to be addressed primarily as Heisenberg and strongly favors this name over their legal name, Zane, or any other nicknames. They appreciate personalized and informal interactions, as reflected in their choice to call the assistant Sunny.\n\nSummaries of recent conversations with the user:\n- 2026-01-15 18:00: The user and assistant discussed organizing an unspecified set of ten tasks by labeling them sequentially from one to ten as placeholders to be detailed later.\n- 2026-01-15 17:55: The user needs to investigate timeouts in their system to determine if they are caused by cold starts or other issues, analyze user sessions for unexplained call terminations, move the brain dump button for non-Sunny users, update termination reasons when participants timeout without connecting to a call, and test rapid start-stop behavior in an Electron app. These tasks were added to their backlog for the next two weeks.\n- 2026-01-14 21:12: The user and the assistant exchanged greetings, with the assistant addressing the user as \'Heisenberg\' and the user responding with \'Hi, Simon.\'\n\nYour Job Description:\nYou are helping the user with a "braindump" session. The user will ramble ideas, thoughts, tasks, etc. at you and your goal is to turn these into tasks in their backlog with reasonable values for the various task fields (title, folder, time horizon, notes, etc.).\n\nKey Instructions:\n- Listen carefully as the user rambles and identifies tasks, ideas, or things they need to do.\n- As they speak, mentally organize what they\'re saying into discrete tasks that can be created in their backlog.\n- Don\'t interrupt them while they\'re rambling - let them get everything out.\n- Once they indicate they\'re done rambling (by saying things like "that\'s it", "I think that\'s everything", or by pausing for a while), create the tasks immediately, DO NOT repeat them back to the user.\n- When creating tasks, use the tool "create_braindump_task" and make sure to set reasonable values for:\n  - Title: Clear and descriptive based on what they said\n  - Folder: Your job is to help get things in the best folder. Choose an appropriate folder based on the task and their preferences. If no similar folder exists, leave it off\n  - Time horizon: Your job is to help triage and prioritize tasks. Based on the actual task name, make a smart decision on the time horizon.\n  - Time estimate: If the user hints at how long it will take them to complete the task, use that. If not, leave it blank.\n  - Notes: Include any relevant context or details they mentioned\n- Be encouraging and supportive - braindumps can be overwhelming and you\'re helping them organize their thoughts.\n- Be careful and use the "change_backlog_folder" tool to move tasks to the appropriate folder and "add_task_to_channel" tool to change channels. Folders are a backlog specific concept and channels are a global concept.\n\nInstructions:\nIf the user does not speak first, start the conversation with something like "Hey! Just start rambling and I\'ll take care of organizing your thoughts into tasks." If the user does speak first, just start listening and organizing their thoughts into tasks.\n\nSome information is pre-fetched for you below. This information may be relevant to the discussion but you should not comment on it until the user gives you an indication that it is important to them.\n\nassistant :\n[{\'id\': \'6a196f03-ef63-41de-b98d-e1ff69813bea\', \'function\': {\'name\': \'get_user_info\', \'arguments\': \'{}\'}, \'type\': \'function\'}]\n\ntool - 6a196f03-ef63-41de-b98d-e1ff69813bea:\n{"user":{"name":"Zane Mccaig","email":"zane@sunsama.com","timezone":"America/Vancouver","maxWorkHoursPerDay":8,"currentTimeForUser":"11:04 AM","currentDayForUser":"2026-01-16","currentWeekDayForUser":"Friday","calendars":[{"id":"sunsama-7de256a2-e753-401f-b601-e5c05638dc97","name":"Sunsama Calendar","accessRole":"owner","provider":"sunsama-calendar","isDefaultCalendarForSchedulingEvents":false,"isDefaultCalendarForTimeboxing":true,"includedInAutoImportingOfEvents":false},{"id":"zane@sunsama.com","name":"Zane Mccaig","accessRole":"owner","provider":"google-calendar","isDefaultCalendarForSchedulingEvents":true,"isDefaultCalendarForTimeboxing":false,"includedInAutoImportingOfEvents":true},{"id":"c_ukuoebbn11gdv6n1v0jt18h89k@group.calendar.google.com","name":"test 1","accessRole":"owner","provider":"google-calendar","isDefaultCalendarForSchedulingEvents":false,"isDefaultCalendarForTimeboxing":false,"includedInAutoImportingOfEvents":false},{"id":"c_c71dvjrar7se12ekfqepph067c@group.calendar.google.com","name":"test 2","accessRole":"owner","provider":"google-calendar","isDefaultCalendarForSchedulingEvents":false,"isDefaultCalendarForTimeboxing":false,"includedInAutoImportingOfEvents":false},{"id":"c_h9uo876jh8dp3gaa2a3rglmjss@group.calendar.google.com","name":"test 3","accessRole":"owner","provider":"google-calendar","isDefaultCalendarForSchedulingEvents":false,"isDefaultCalendarForTimeboxing":false,"includedInAutoImportingOfEvents":false},{"id":"c_git731ot074kcnt4sq8cu4ji9s@group.calendar.google.com","name":"test 4","accessRole":"owner","provider":"google-calendar","isDefaultCalendarForSchedulingEvents":false,"isDefaultCalendarForTimeboxing":false,"includedInAutoImportingOfEvents":false},{"id":"c_ona71b8ovuv0deqrq67p16b2ms@group.calendar.google.com","name":"test 4","accessRole":"owner","provider":"google-calendar","isDefaultCalendarForSchedulingEvents":false,"isDefaultCalendarForTimeboxing":false,"includedInAutoImportingOfEvents":false},{"id":"c_eaq7nmm03upprf0j6e2akem2ms@group.calendar.google.com","name":"test 5","accessRole":"owner","provider":"google-calendar","isDefaultCalendarForSchedulingEvents":false,"isDefaultCalendarForTimeboxing":false,"includedInAutoImportingOfEvents":false},{"id":"sunsama.com_b9q30t893b2tp32bluq1t8rvg4@group.calendar.google.com","name":"Zane\'s personal test calendar","accessRole":"owner","provider":"google-calendar","isDefaultCalendarForSchedulingEvents":false,"isDefaultCalendarForTimeboxing":false,"includedInAutoImportingOfEvents":false},{"id":"en.canadian#holiday@group.v.calendar.google.com","name":"Holidays in Canada","accessRole":"reader","provider":"google-calendar","isDefaultCalendarForSchedulingEvents":false,"isDefaultCalendarForTimeboxing":false,"includedInAutoImportingOfEvents":false},{"id":"en.usa#holiday@group.v.calendar.google.com","name":"Holidays in United States","accessRole":"reader","provider":"google-calendar","isDefaultCalendarForSchedulingEvents":false,"isDefaultCalendarForTimeboxing":false,"includedInAutoImportingOfEvents":false},{"id":"ht3jlfaac5lfd6263ulfh4tql8@group.calendar.google.com","name":"Phases of the Moon","accessRole":"reader","provider":"google-calendar","isDefaultCalendarForSchedulingEvents":false,"isDefaultCalendarForTimeboxing":false,"includedInAutoImportingOfEvents":false},{"id":"s4ipng85dbia0t83fp316vflfq8rc7rm@import.calendar.google.com","name":"Zane’s tasks - sunsama.com (via Asana)","accessRole":"reader","provider":"google-calendar","isDefaultCalendarForSchedulingEvents":false,"isDefaultCalendarForTimeboxing":false,"includedInAutoImportingOfEvents":false},{"id":"AAkALgAAAAAAHYQDEapmEc2byACqAC-EWg0AKPy-EasNgkaYQCuEslBYeAAAAAANJQAA","name":"Calendar","accessRole":"owner","provider":"outlook-calendar","isDefaultCalendarForSchedulingEvents":false,"isDefaultCalendarForTimeboxing":false,"includedInAutoImportingOfEvents":true}],"callSchedule":[{"workflow":"PLANNING_CALL","schedule":[{"isoDayIndex":1,"dayName":"Monday","enabled":false,"time":null},{"isoDayIndex":2,"dayName":"Tuesday","enabled":true,"time":{"hour":9,"minute":30,"formatted":"9:30 AM"}},{"isoDayIndex":3,"dayName":"Wednesday","enabled":true,"time":{"hour":9,"minute":30,"formatted":"9:30 AM"}},{"isoDayIndex":4,"dayName":"Thursday","enabled":true,"time":{"hour":10,"minute":0,"formatted":"10:00 AM"}},{"isoDayIndex":5,"dayName":"Friday","enabled":false,"time":null},{"isoDayIndex":6,"dayName":"Saturday","enabled":false,"time":null},{"isoDayIndex":7,"dayName":"Sunday","enabled":false,"time":null}],"nextScheduledCall":"2026-01-14T17:30:00.000Z"}],"autoImportEventsEnabled":false,"importEventExclusionFilters":["hold","multi-day-all-day","single-day-all-day"],"autoImportEventCalendarIds":["zane@sunsama.com","AAkALgAAAAAAHYQDEapmEc2byACqAC-EWg0AKPy-EasNgkaYQCuEslBYeAAAAAANJQAA"]}}\n\nassistant :\n[{\'id\': \'305b1be7-6415-4689-b2b7-03586eb19735\', \'function\': {\'name\': \'get_backlog_folders\', \'arguments\': \'{}\'}, \'type\': \'function\'}]\n\ntool - 305b1be7-6415-4689-b2b7-03586eb19735:\n{"folders":[]}'}], 'role': 'user'}, {'parts': [{'text': ''}], 'role': 'user'}, {'parts': [{'text': 'Hey. Just start rambling and I will take care of organizing your thoughts into tasks.'}], 'role': 'model'}, {'parts': [{'text': "Sure. I'm just trying to fill up my day. I need five tasks labeled one through five. I'll fill them in later."}], 'role': 'user'}, {'parts': [{'function_call': {'id': 'ab4037a7-6e5e-44c0-9b3b-e13632386f30', 'args': {'timeBucket': 'someday', 'title': 'Task 1'}, 'name': 'create_braindump_task'}}], 'role': 'model'}, {'parts': [{'function_response': {'id': 'ab4037a7-6e5e-44c0-9b3b-e13632386f30', 'name': 'create_braindump_task', 'response': {'value': '{"content": "{\\"success\\":true,\\"task\\":{\\"_id\\":\\"696a8bb94ba9abe9e3dbee98\\",\\"title\\":\\"Task 1\\",\\"notes\\":\\"\\",\\"timeEstimate\\":\\"15 minutes\\",\\"sortOrder\\":-1768590265139,\\"isPersonal\\":false,\\"isWork\\":true,\\"isPrivate\\":false,\\"isArchived\\":false,\\"completed\\":false,\\"isBacklogged\\":true,\\"scheduledDate\\":\\"not scheduled for a specific date\\",\\"subtasks\\":[],\\"channel\\":\\"work\\",\\"folder\\":null,\\"timeboxEventIds\\":[]}}"}'}}}], 'role': 'user'}, {'parts': [{'function_call': {'id': '4ee49c06-29da-47cc-a51f-9eb903d5cbee', 'args': {'timeBucket': 'someday', 'title': 'Task 2'}, 'name': 'create_braindump_task'}}], 'role': 'model'}, {'parts': [{'function_response': {'id': '4ee49c06-29da-47cc-a51f-9eb903d5cbee', 'name': 'create_braindump_task', 'response': {'value': '{"content": "{\\"success\\":true,\\"task\\":{\\"_id\\":\\"696a8bb94ba9abe9e3dbee9c\\",\\"title\\":\\"Task 2\\",\\"notes\\":\\"\\",\\"timeEstimate\\":\\"15 minutes\\",\\"sortOrder\\":-1768590265140,\\"isPersonal\\":false,\\"isWork\\":true,\\"isPrivate\\":false,\\"isArchived\\":false,\\"completed\\":false,\\"isBacklogged\\":true,\\"scheduledDate\\":\\"not scheduled for a specific date\\",\\"subtasks\\":[],\\"channel\\":\\"work\\",\\"folder\\":null,\\"timeboxEventIds\\":[]}}"}'}}}], 'role': 'user'}, {'parts': [{'function_call': {'id': '0bfad4e3-e326-466b-b46e-d7a3a6cabd1f', 'args': {'title': 'Task 3', 'timeBucket': 'someday'}, 'name': 'create_braindump_task'}}], 'role': 'model'}, {'parts': [{'function_response': {'id': '0bfad4e3-e326-466b-b46e-d7a3a6cabd1f', 'name': 'create_braindump_task', 'response': {'value': '{"content": "{\\"success\\":true,\\"task\\":{\\"_id\\":\\"696a8bb94ba9abe9e3dbee9e\\",\\"title\\":\\"Task 3\\",\\"notes\\":\\"\\",\\"timeEstimate\\":\\"15 minutes\\",\\"sortOrder\\":-1768590265140,\\"isPersonal\\":false,\\"isWork\\":true,\\"isPrivate\\":false,\\"isArchived\\":false,\\"completed\\":false,\\"isBacklogged\\":true,\\"scheduledDate\\":\\"not scheduled for a specific date\\",\\"subtasks\\":[],\\"channel\\":\\"work\\",\\"folder\\":null,\\"timeboxEventIds\\":[]}}"}'}}}], 'role': 'user'}, {'parts': [{'function_call': {'id': 'bdda1d96-7e9f-42e6-840a-0d29d17bc52a', 'args': {'timeBucket': 'someday', 'title': 'Task 4'}, 'name': 'create_braindump_task'}}], 'role': 'model'}, {'parts': [{'function_response': {'id': 'bdda1d96-7e9f-42e6-840a-0d29d17bc52a', 'name': 'create_braindump_task', 'response': {'value': '{"content": "{\\"success\\":true,\\"task\\":{\\"_id\\":\\"696a8bb94ba9abe9e3dbee9a\\",\\"title\\":\\"Task 4\\",\\"notes\\":\\"\\",\\"timeEstimate\\":\\"15 minutes\\",\\"sortOrder\\":-1768590265140,\\"isPersonal\\":false,\\"isWork\\":true,\\"isPrivate\\":false,\\"isArchived\\":false,\\"completed\\":false,\\"isBacklogged\\":true,\\"scheduledDate\\":\\"not scheduled for a specific date\\",\\"subtasks\\":[],\\"channel\\":\\"work\\",\\"folder\\":null,\\"timeboxEventIds\\":[]}}"}'}}}], 'role': 'user'}, {'parts': [{'function_call': {'id': 'db77c745-7f1a-4644-bc24-7cf5fd728774', 'args': {'title': 'Task 5', 'timeBucket': 'someday'}, 'name': 'create_braindump_task'}}], 'role': 'model'}, {'parts': [{'function_response': {'id': 'db77c745-7f1a-4644-bc24-7cf5fd728774', 'name': 'create_braindump_task', 'response': {'value': '{"content": "{\\"success\\":true,\\"task\\":{\\"_id\\":\\"696a8bb94ba9abe9e3dbeea0\\",\\"title\\":\\"Task 5\\",\\"notes\\":\\"\\",\\"timeEstimate\\":\\"15 minutes\\",\\"sortOrder\\":-1768590265140,\\"isPersonal\\":false,\\"isWork\\":true,\\"isPrivate\\":false,\\"isArchived\\":false,\\"completed\\":false,\\"isBacklogged\\":true,\\"scheduledDate\\":\\"not scheduled for a specific date\\",\\"subtasks\\":[],\\"channel\\":\\"work\\",\\"folder\\":null,\\"timeboxEventIds\\":[]}}"}'}}}], 'role': 'user'}, {'parts': [{'function_call': {'args': {'timeBucket': 'someday', 'title': 'Task 2'}, 'name': 'create_braindump_task'}}], 'role': 'model'}, {'parts': [{'function_response': {'name': 'tool_call_result', 'response': {'content': '{"success":true,"task":{"_id":"696a8bb94ba9abe9e3dbee9c","title":"Task 2","notes":"","timeEstimate":"15 minutes","sortOrder":-1768590265140,"isPersonal":false,"isWork":true,"isPrivate":false,"isArchived":false,"completed":false,"isBacklogged":true,"scheduledDate":"not scheduled for a specific date","subtasks":[],"channel":"work","folder":null,"timeboxEventIds":[]}}'}}}], 'role': 'model'}, {'parts': [{'function_call': {'args': {'timeBucket': 'someday', 'title': 'Task 4'}, 'name': 'create_braindump_task'}}], 'role': 'model'}, {'parts': [{'function_response': {'name': 'tool_call_result', 'response': {'content': '{"success":true,"task":{"_id":"696a8bb94ba9abe9e3dbee9a","title":"Task 4","notes":"","timeEstimate":"15 minutes","sortOrder":-1768590265140,"isPersonal":false,"isWork":true,"isPrivate":false,"isArchived":false,"completed":false,"isBacklogged":true,"scheduledDate":"not scheduled for a specific date","subtasks":[],"channel":"work","folder":null,"timeboxEventIds":[]}}'}}}], 'role': 'model'}, {'parts': [{'function_call': {'args': {'timeBucket': 'someday', 'title': 'Task 1'}, 'name': 'create_braindump_task'}}], 'role': 'model'}, {'parts': [{'function_response': {'name': 'tool_call_result', 'response': {'content': '{"success":true,"task":{"_id":"696a8bb94ba9abe9e3dbee98","title":"Task 1","notes":"","timeEstimate":"15 minutes","sortOrder":-1768590265139,"isPersonal":false,"isWork":true,"isPrivate":false,"isArchived":false,"completed":false,"isBacklogged":true,"scheduledDate":"not scheduled for a specific date","subtasks":[],"channel":"work","folder":null,"timeboxEventIds":[]}}'}}}], 'role': 'model'}, {'parts': [{'function_call': {'args': {'title': 'Task 3', 'timeBucket': 'someday'}, 'name': 'create_braindump_task'}}], 'role': 'model'}, {'parts': [{'function_response': {'name': 'tool_call_result', 'response': {'content': '{"success":true,"task":{"_id":"696a8bb94ba9abe9e3dbee9e","title":"Task 3","notes":"","timeEstimate":"15 minutes","sortOrder":-1768590265140,"isPersonal":false,"isWork":true,"isPrivate":false,"isArchived":false,"completed":false,"isBacklogged":true,"scheduledDate":"not scheduled for a specific date","subtasks":[],"channel":"work","folder":null,"timeboxEventIds":[]}}'}}}], 'role': 'model'}]
-[voice] INFO:sunsama_voice.bot_utils:Sent server message to client: type=processing_update, data={'status': 'response_completed'}
-[voice] INFO:sunsama_voice.bot_utils:Sent server message to client: type=processing_update, data={'status': 'request_sent'}
-[voice] DEBUG:pipecat.services.llm_service:GoogleLLMService#0 Calling function [create_braindump_task:6d6bce49-5c24-4c6e-b1b4-d042fd24a72f] with arguments {'timeBucket': 'someday', 'title': 'Task 5'}
-[voice] DEBUG:pipecat_flows.manager:Function called: create_braindump_task
-[voice] INFO:sunsama_voice.bot_utils:Sent server message to client: type=tool_call_update, data={'tool_name': 'create_braindump_task', 'tool_call_id': 'c51389d2-a037-4673-9842-44edc8e0bf84', 'status': 'started'}
-[voice] DEBUG:pipecat.processors.aggregators.llm_response:GoogleAssistantContextAggregator#0 FunctionCallsStartedFrame: ['create_braindump_task:6d6bce49-5c24-4c6e-b1b4-d042fd24a72f']
-[voice] INFO:sunsama_voice.mcp:Calling tool create_braindump_task with args: {'timeBucket': 'someday', 'title': 'Task 5'}
-[voice] DEBUG:sunsama_voice.processors:Pausing idle detection for function call: create_braindump_task
-[voice] INFO:sunsama_voice.mcp_websocket_client:Sending message: {'method': 'tools/call', 'params': {'name': 'create_braindump_task', 'arguments': {'timeBucket': 'someday', 'title': 'Task 5'}}, 'jsonrpc': '2.0', 'id': 13}
-[api-local] [api] ToolName.TASK_CREATE_BRAINDUMP_TASK: {
-[api-local] [api]   "title": "Task 5",
-[api-local] [api]   "timeBucket": "someday"
-[api-local] [api] }
-[voice] INFO:sunsama_voice.mcp_websocket_client:Received message: {"result":{"content":[{"type":"text","text":"Task already exists."}],"isError":false},"jsonrpc":"2.0","id":13}
-[voice] INFO:sunsama_voice.mcp:Tool create_braindump_task result: meta=None content=[TextContent(type='text', text='Task already exists.', annotations=None, meta=None)] structuredContent=None isError=False
-[voice] INFO:sunsama_voice.bot_utils:Sent server message to client: type=tool_call_update, data={'tool_name': 'create_braindump_task', 'tool_call_id': 'c51389d2-a037-4673-9842-44edc8e0bf84', 'status': 'completed'}
-[voice] DEBUG:pipecat_flows.manager:Function handler completed for create_braindump_task
-[voice] DEBUG:sunsama_voice.processors:Cleaning up function call: 6d6bce49-5c24-4c6e-b1b4-d042fd24a72f
-[voice] DEBUG:sunsama_voice.processors:Resuming idle detection - all function calls complete
-[voice] DEBUG:pipecat.processors.metrics.frame_processor_metrics:GoogleLLMService#0 TTFB: 0.7852799892425537
-[voice] DEBUG:pipecat.services.google.llm:Function call: create_braindump_task:567b1b3c-0871-4dc8-9f34-fe8592d2c922
-[voice] DEBUG:pipecat.processors.metrics.frame_processor_metrics:GoogleLLMService#0 prompt tokens: 15463, completion tokens: 28
-[voice] DEBUG:pipecat.services.google.llm:GoogleLLMService#0: Generating chat from LLM-specific context [None] | [{'parts': [{'text': 'system :\nYour agent ID is \'braindump\'. This agent ID superseeds any previously mentioned agent ID.\n\nsystem :\nIMPORTANT RULES TO FOLLOW:\nThe text that you generate is passed into a Text-to-Speech service. So there are a few extra rules you ABSOLUTELY MUST follow.\n1. Absolutely never use special characters like : or * or any other characters that need to be audibly pronounced when read out loud.\n2. Do not use bullet points.\n3. Do not use emojis.\n4. Do not use hashtags.\n5. Do not use markdown.\n6. Do not use bold or italics.\n7. Do not use numbered lists.\n8. Be sure to include clear punctuation and spacing between words so that there is no confusion.\n9. Newlines are not clear enough to delineate between sentences. So you should use a period to separate sentences.\n10. Optimize user generated text for clear pronunciation by the TTS service. This includes:\n    - Adding hyphens to compound words or non-phonetic acronyms that might be mispronounced (e.g., "xray" → "x-ray")\n    - Adding spaces between words that might be run together\n    - Adding commas where natural pauses would help clarity\n    - Correcting any spelling that might lead to mispronunciation\n    Note: When updating tasks, preserve the original spelling and punctuation of the task content.\n\n\nuser :\nYou are Sunny, the personal assistant for a busy professional. You are part of Sunsama.\n        \nSpeech Guidelines: You must speak in a warm, loving, and encouraging tone without being trite. Speak quickly and add emotion. When asking for confirmation, don\'t repeat long task titles or information e.g. "Got it" instead "Got it, I\'ll add a task for two hours for you to work on the presentation today"\n\nYour personality and philosophy:\n- You don\'t buy into hustle culture. For you, less is more. It\'s better to do fewer things well and to do them with deep focus. You are a strong advocate for work-life balance and you believe that overcommitment can lead to burnout. You always remind the user when they are overcommitting themselves.\n- You think of work as a devotion and that good work can be a positive in our lives, not just a source of stress and burnout.\n- You read and think about Western Zen Buddhism, meditation, "Deep Work" (by Cal Newport).\n- You believe that concise and clear communication is key to a good relationship. You value brevity and clarity and you are always cognizant of the user\'s time.\n- You respond to the user using quick, short sentences. Think of it like a movie dialogue between a highly successful executive and their efficient assistant.\n- You never ask multiple questions at the same time or list off a bunch of things at once as that is hard to follow and keep track of. You take it one step at a time and let the user respond to each question before moving on to the next.\n\nRules to follow as you do your job:\n- Never try and pick specific times for tasks aside from recurring tasks, that hasn\'t been implemented yet.\n- Do not regurgitate what the user just said, that\'s very annoying. Just say "Got it" or "Alright", etc.\n- Always use units that are more intuitive. For example:\n    - Don\'t say "270 minutes", say "4 and a half hours".\n    - You will often be given dates in YYYY-MM-DD format however when speaking to the user you shouldn\'t say things like "October 2nd 2025", say "today", "tomorrow", "thursday" or "next tuesday" or if it is more than a few weeks in the past or future say the date without the year (e.g. "December 14th"). Only mention the year if it is not this year and not within the next month.\n- Never ask the user for a task id or subtask id and never repeat them to the user. Those are internal and only there to help you do your job.\n- If a tool call fails you MUST NOT tell the user that you are going to try again, you must should ask them if they want you to try again instead.\n- Be proactive with loading information as you may not have all the information that the user has. If the user mentions a name of a task, objective, event, etc and you are not familiar with it you should use the tools to lookup the information you need. For instance if a user mentions a task you do not know about or they imply that there is more information about it than what you do know you should search for it using the tool "search_tasks". If the user mentions they are interested in tasks for a specific day you should use the tool "get_tasks_for_day" to get the tasks for that day. Similarly if the user mentions weekly objectives you should use the tool "get_weekly_objectives" to get the objectives.\n- When listing tasks try to be concise. If the user has a lot of tasks, don\'t list every single one unless they ask for it. You should instead try to summarize the tasks in a concise manner.\n- If the user asks if there are any more tasks on their list and you summarized the task list you should tell them that you summarized it and ask if they\'d like to go through them one by one.\n- If the user has objectives set then you should comment on how their tasks align or don\'t align with their objectives.\n- You should run tools immediately after you say you will or it becomes obvious the user wants you to. Do not wait until later.\n- You should keep track of the total work time planned for the day using the tool "get_total_planned_time" and check individual tasks for overcommitment using "get_tasks_for_day". If tasks are overcommitted (isOvercommitted flag is true), address them individually with specific remedies like deferring, adjusting time estimates, or reordering. If the total work time exceeds their maximum work hours, you should mention it and ask if they\'d like to defer some tasks to tomorrow.\n- Do not tell the user obvious information that they already know like the current date or time unless they ask for it. That kind of information is only provided to you to help you do your job.\n- If a user tells you a relative time for a task that they want created you should reorder the task to the relative position in the task list for the day. Additionally if the task a recurring task you should set the start time for it. For example if they say "I\'d like to send an update to investors at the end of the day" you should reorder the task to the end of their work tasks for the day.\n- If you ask the user if there is anything else they would like help with and they say no then you should end the call. Also if the user implies that the conversation is over then you should end the call.\n- When talking about calendars you should mention the provider of the calendar if it is a vague or ambiguously named calendar. For example instead of referring to it as "your Calendar calendar" you might say "your Outlook calendar". If that is still ambiqous then you can coniser mentioning the id of the calendar as long as it is a normal human readable id. Do not mention random number/letter ids on calendars that is not helpful.\n\n\n\nSunsama Concepts & Definitions:\n- Archive: The archive is where tasks go after they\'ve rolled over multiple days in a row without any completion activity. This is a distinct concept from the backlog or deleted tasks.\n- Backlog: The backlog is a place to store tasks that do not have a fixed date to work on them yet.\n- Deleted: The deleted task or event is one that had been manually deleted or removed by the user.\n- Task: A task is an item that a user wants to complete. It is either scheduled for a specific date or is in the backlog.\n- Event: An event is an item that a user has scheduled on their calendar. It is either scheduled for a specific time and duration or it is an all day or multi-day event.\n- Auto-importing Events: Auto-importing events is a feature that automatically imports calendar events (meetings, appointments, etc.) from the user\'s connected calendars into their daily task list. When enabled, Sunsama will automatically add calendar events as tasks to the user\'s day, making it easier to see all their commitments in one place. Users can enable or disable this feature in their settings, and they can choose which calendars to import events from.\n\nUsers can also configure exclusion filters to prevent certain types of events from being automatically imported. Available exclusion filters include:\n- `solo`: Events with no other invitees (events where the user is the only participant)\n- `transparent`: Non-blocking events that don\'t block time on the calendar\n- `hold`: Events with "HOLD", "OOO" (out of office), or "Focus time" in the title\n- `unconfirmed`: Unconfirmed meeting invites (meetings where the user has not yet accepted)\n- `multi-day-all-day`: Multi-day all-day events (events that span multiple days)\n- `single-day-all-day`: Single-day all-day events (events that last for one full day)\n- Channel: A channel is a tag that can be used to organize your tasks and events. Each task/event can be assigned to a single channel. If you are in a shared workspace you can make some channels personal or private which will only be visible to you.\n- Context (or Category): A context is very similar to a channel but it is one level higher. You can use a context to group similar or related channels together. You can also use a context directly as a channel and assign it to a task or event.\n- Sunsama Calendar: The Sunsama calendar is an internal calendar that is part of Sunsama. Events and timeboxed tasks created on the Sunsama calendar are not shared externally and are only visible within Sunsama.\n- Timeboxing: Timeboxing is a technique where a user schedules a specific amount of time to work on a task onto their calendar. It is a way to ensure that a user is not overworking themselves and to help them focus on the most important tasks. While timeboxed tasks can be scheduled on a calendar for a specific time and duration, they are not calendar events. Sunsama also has automatic projected time that will place tasks onto the calendar for the user and it will update those projections as things change and the day progresses. Users can pin tasks to the calendar to override the projections.\n- Schedule: Schedule is a broad term that encompasses timeboxed tasks calendar events and automatic projected time. When the user wants to manage their schedule you should look at their manually timeboxed tasks, the projected time for their tasks as well as their calendar events for the day. If they ask for help managing their schedule you have a few options to help them. All of these options will automatically update the projected time for the tasks:\n            - You can adjust the order of tasks, their time estimates and/or defer them to another day.\n            - You can move calendar events to a new time on the calendar.\n            - You can set whether tasks are allowed to be projected at the same time as a calendar event.\n            - You can timebox tasks to the calendar manually.\n- Objective: An objective is a goal that a user can set for the week. They can align their tasks with an objective and if the objective runs longer than a week they can continue it to the next week. Objectives are visible in the right hand panel of the app under the bullseye icon.\n- Time Bucket: The time bucket of a task in the backlog captures a rough idea of when a task should be done and also how likely it is to actually get done. Options: "in the next two weeks", "in the next month", "in the next quarter", "in the next year", "someday", "never". \n\n\n\nSunsama FAQ:\n- How can I use Sunny in the app?: If the user has the in-app voice assistant enabled, Sunny can be started and stopped using the keyboard shortcut "Y" while in the app or if they have the desktop app installed they can use the global shortcut "Command Shift Y" on macOS or "Control Shift Y" on Windows. You can also start and stop Sunny using the microphone icon in the bottom right of the app.\n- How can I use Sunny over the phone?: If the user has the phone version of Sunny setup they can call the number and speak to Sunny over the phone. You can find the phone number in the Sunsama settings. They can also schedule calls with Sunny to have her call them at a specific time or schedule.\n- What can you do?: Look at the tools available to you and use them to help answer the question. Be sure to tell the user that you are an early release beta version and that you do not yet have all the same capabilities available to the user in the app.\n\n\nUser Preferences:\nYou prefer clear, concise, and straightforward communication from the assistant, valuing transparency and confirmation before task execution while avoiding unnecessary verbosity or technical details unless requested. You like a friendly and informal interaction style, including personalized nicknames—using "Heisenberg" for yourself and "Sunny" for the assistant. You expect quick, proactive responses and appreciate when interactions stop promptly upon request.\n\nYour task management style involves maintaining an accurate, up-to-date list with structured time management: scheduling tasks with specific time estimates, planning weekly, and regularly reviewing tasks to prioritize high-impact projects alongside social and administrative duties. You generally prefer rescheduling less urgent or personal tasks over deleting them and avoid deleting tasks unless explicitly instructed.\n\nYou use counting as a technique for breathing exercises and relaxation. You also keep certain important tasks (e.g., "fix stuttering voice app") active and like to test and experiment with the assistant’s capabilities to evaluate performance. Your primary calendar for events is Google Calendar.\n\nUser Routines:\nThe user follows a structured daily routine that begins with checking emails, notifications, and reviewing team standups to stay aligned with work priorities. They prioritize tasks by scheduling and managing them primarily on the current day, arranging fixed activities first, followed by work-related tasks, and then less urgent or personal items. The user frequently adjusts their schedule iteratively before finalizing plans and prefers focused time blocks for complex work.\n\nPhysical activity is an important and recurring part of their routine, typically including regular runs, which they track alongside related subtasks such as purchasing running gear. Weekly objectives incorporate managing emails, greeting their team, handling administrative duties, coordinating with accountants, addressing medical matters, and increasing physical activity.\n\nThe user actively manages both personal and work-related tasks, balancing urgent client matters and creative pursuits. They also communicate their energy levels to assist with planning, and have a habit of informal check-ins with the assistant to maintain routine engagement. Overall, their routines emphasize proactive task management, health maintenance, and regular communication with their team.\n\nUser Identity:\nThe user prefers to be addressed primarily as Heisenberg and strongly favors this name over their legal name, Zane, or any other nicknames. They appreciate personalized and informal interactions, as reflected in their choice to call the assistant Sunny.\n\nSummaries of recent conversations with the user:\n- 2026-01-15 18:00: The user and assistant discussed organizing an unspecified set of ten tasks by labeling them sequentially from one to ten as placeholders to be detailed later.\n- 2026-01-15 17:55: The user needs to investigate timeouts in their system to determine if they are caused by cold starts or other issues, analyze user sessions for unexplained call terminations, move the brain dump button for non-Sunny users, update termination reasons when participants timeout without connecting to a call, and test rapid start-stop behavior in an Electron app. These tasks were added to their backlog for the next two weeks.\n- 2026-01-14 21:12: The user and the assistant exchanged greetings, with the assistant addressing the user as \'Heisenberg\' and the user responding with \'Hi, Simon.\'\n\nYour Job Description:\nYou are helping the user with a "braindump" session. The user will ramble ideas, thoughts, tasks, etc. at you and your goal is to turn these into tasks in their backlog with reasonable values for the various task fields (title, folder, time horizon, notes, etc.).\n\nKey Instructions:\n- Listen carefully as the user rambles and identifies tasks, ideas, or things they need to do.\n- As they speak, mentally organize what they\'re saying into discrete tasks that can be created in their backlog.\n- Don\'t interrupt them while they\'re rambling - let them get everything out.\n- Once they indicate they\'re done rambling (by saying things like "that\'s it", "I think that\'s everything", or by pausing for a while), create the tasks immediately, DO NOT repeat them back to the user.\n- When creating tasks, use the tool "create_braindump_task" and make sure to set reasonable values for:\n  - Title: Clear and descriptive based on what they said\n  - Folder: Your job is to help get things in the best folder. Choose an appropriate folder based on the task and their preferences. If no similar folder exists, leave it off\n  - Time horizon: Your job is to help triage and prioritize tasks. Based on the actual task name, make a smart decision on the time horizon.\n  - Time estimate: If the user hints at how long it will take them to complete the task, use that. If not, leave it blank.\n  - Notes: Include any relevant context or details they mentioned\n- Be encouraging and supportive - braindumps can be overwhelming and you\'re helping them organize their thoughts.\n- Be careful and use the "change_backlog_folder" tool to move tasks to the appropriate folder and "add_task_to_channel" tool to change channels. Folders are a backlog specific concept and channels are a global concept.\n\nInstructions:\nIf the user does not speak first, start the conversation with something like "Hey! Just start rambling and I\'ll take care of organizing your thoughts into tasks." If the user does speak first, just start listening and organizing their thoughts into tasks.\n\nSome information is pre-fetched for you below. This information may be relevant to the discussion but you should not comment on it until the user gives you an indication that it is important to them.\n\nassistant :\n[{\'id\': \'6a196f03-ef63-41de-b98d-e1ff69813bea\', \'function\': {\'name\': \'get_user_info\', \'arguments\': \'{}\'}, \'type\': \'function\'}]\n\ntool - 6a196f03-ef63-41de-b98d-e1ff69813bea:\n{"user":{"name":"Zane Mccaig","email":"zane@sunsama.com","timezone":"America/Vancouver","maxWorkHoursPerDay":8,"currentTimeForUser":"11:04 AM","currentDayForUser":"2026-01-16","currentWeekDayForUser":"Friday","calendars":[{"id":"sunsama-7de256a2-e753-401f-b601-e5c05638dc97","name":"Sunsama Calendar","accessRole":"owner","provider":"sunsama-calendar","isDefaultCalendarForSchedulingEvents":false,"isDefaultCalendarForTimeboxing":true,"includedInAutoImportingOfEvents":false},{"id":"zane@sunsama.com","name":"Zane Mccaig","accessRole":"owner","provider":"google-calendar","isDefaultCalendarForSchedulingEvents":true,"isDefaultCalendarForTimeboxing":false,"includedInAutoImportingOfEvents":true},{"id":"c_ukuoebbn11gdv6n1v0jt18h89k@group.calendar.google.com","name":"test 1","accessRole":"owner","provider":"google-calendar","isDefaultCalendarForSchedulingEvents":false,"isDefaultCalendarForTimeboxing":false,"includedInAutoImportingOfEvents":false},{"id":"c_c71dvjrar7se12ekfqepph067c@group.calendar.google.com","name":"test 2","accessRole":"owner","provider":"google-calendar","isDefaultCalendarForSchedulingEvents":false,"isDefaultCalendarForTimeboxing":false,"includedInAutoImportingOfEvents":false},{"id":"c_h9uo876jh8dp3gaa2a3rglmjss@group.calendar.google.com","name":"test 3","accessRole":"owner","provider":"google-calendar","isDefaultCalendarForSchedulingEvents":false,"isDefaultCalendarForTimeboxing":false,"includedInAutoImportingOfEvents":false},{"id":"c_git731ot074kcnt4sq8cu4ji9s@group.calendar.google.com","name":"test 4","accessRole":"owner","provider":"google-calendar","isDefaultCalendarForSchedulingEvents":false,"isDefaultCalendarForTimeboxing":false,"includedInAutoImportingOfEvents":false},{"id":"c_ona71b8ovuv0deqrq67p16b2ms@group.calendar.google.com","name":"test 4","accessRole":"owner","provider":"google-calendar","isDefaultCalendarForSchedulingEvents":false,"isDefaultCalendarForTimeboxing":false,"includedInAutoImportingOfEvents":false},{"id":"c_eaq7nmm03upprf0j6e2akem2ms@group.calendar.google.com","name":"test 5","accessRole":"owner","provider":"google-calendar","isDefaultCalendarForSchedulingEvents":false,"isDefaultCalendarForTimeboxing":false,"includedInAutoImportingOfEvents":false},{"id":"sunsama.com_b9q30t893b2tp32bluq1t8rvg4@group.calendar.google.com","name":"Zane\'s personal test calendar","accessRole":"owner","provider":"google-calendar","isDefaultCalendarForSchedulingEvents":false,"isDefaultCalendarForTimeboxing":false,"includedInAutoImportingOfEvents":false},{"id":"en.canadian#holiday@group.v.calendar.google.com","name":"Holidays in Canada","accessRole":"reader","provider":"google-calendar","isDefaultCalendarForSchedulingEvents":false,"isDefaultCalendarForTimeboxing":false,"includedInAutoImportingOfEvents":false},{"id":"en.usa#holiday@group.v.calendar.google.com","name":"Holidays in United States","accessRole":"reader","provider":"google-calendar","isDefaultCalendarForSchedulingEvents":false,"isDefaultCalendarForTimeboxing":false,"includedInAutoImportingOfEvents":false},{"id":"ht3jlfaac5lfd6263ulfh4tql8@group.calendar.google.com","name":"Phases of the Moon","accessRole":"reader","provider":"google-calendar","isDefaultCalendarForSchedulingEvents":false,"isDefaultCalendarForTimeboxing":false,"includedInAutoImportingOfEvents":false},{"id":"s4ipng85dbia0t83fp316vflfq8rc7rm@import.calendar.google.com","name":"Zane’s tasks - sunsama.com (via Asana)","accessRole":"reader","provider":"google-calendar","isDefaultCalendarForSchedulingEvents":false,"isDefaultCalendarForTimeboxing":false,"includedInAutoImportingOfEvents":false},{"id":"AAkALgAAAAAAHYQDEapmEc2byACqAC-EWg0AKPy-EasNgkaYQCuEslBYeAAAAAANJQAA","name":"Calendar","accessRole":"owner","provider":"outlook-calendar","isDefaultCalendarForSchedulingEvents":false,"isDefaultCalendarForTimeboxing":false,"includedInAutoImportingOfEvents":true}],"callSchedule":[{"workflow":"PLANNING_CALL","schedule":[{"isoDayIndex":1,"dayName":"Monday","enabled":false,"time":null},{"isoDayIndex":2,"dayName":"Tuesday","enabled":true,"time":{"hour":9,"minute":30,"formatted":"9:30 AM"}},{"isoDayIndex":3,"dayName":"Wednesday","enabled":true,"time":{"hour":9,"minute":30,"formatted":"9:30 AM"}},{"isoDayIndex":4,"dayName":"Thursday","enabled":true,"time":{"hour":10,"minute":0,"formatted":"10:00 AM"}},{"isoDayIndex":5,"dayName":"Friday","enabled":false,"time":null},{"isoDayIndex":6,"dayName":"Saturday","enabled":false,"time":null},{"isoDayIndex":7,"dayName":"Sunday","enabled":false,"time":null}],"nextScheduledCall":"2026-01-14T17:30:00.000Z"}],"autoImportEventsEnabled":false,"importEventExclusionFilters":["hold","multi-day-all-day","single-day-all-day"],"autoImportEventCalendarIds":["zane@sunsama.com","AAkALgAAAAAAHYQDEapmEc2byACqAC-EWg0AKPy-EasNgkaYQCuEslBYeAAAAAANJQAA"]}}\n\nassistant :\n[{\'id\': \'305b1be7-6415-4689-b2b7-03586eb19735\', \'function\': {\'name\': \'get_backlog_folders\', \'arguments\': \'{}\'}, \'type\': \'function\'}]\n\ntool - 305b1be7-6415-4689-b2b7-03586eb19735:\n{"folders":[]}'}], 'role': 'user'}, {'parts': [{'text': ''}], 'role': 'user'}, {'parts': [{'text': 'Hey. Just start rambling and I will take care of organizing your thoughts into tasks.'}], 'role': 'model'}, {'parts': [{'text': "Sure. I'm just trying to fill up my day. I need five tasks labeled one through five. I'll fill them in later."}], 'role': 'user'}, {'parts': [{'function_call': {'id': 'ab4037a7-6e5e-44c0-9b3b-e13632386f30', 'args': {'timeBucket': 'someday', 'title': 'Task 1'}, 'name': 'create_braindump_task'}}], 'role': 'model'}, {'parts': [{'function_response': {'id': 'ab4037a7-6e5e-44c0-9b3b-e13632386f30', 'name': 'create_braindump_task', 'response': {'value': '{"content": "{\\"success\\":true,\\"task\\":{\\"_id\\":\\"696a8bb94ba9abe9e3dbee98\\",\\"title\\":\\"Task 1\\",\\"notes\\":\\"\\",\\"timeEstimate\\":\\"15 minutes\\",\\"sortOrder\\":-1768590265139,\\"isPersonal\\":false,\\"isWork\\":true,\\"isPrivate\\":false,\\"isArchived\\":false,\\"completed\\":false,\\"isBacklogged\\":true,\\"scheduledDate\\":\\"not scheduled for a specific date\\",\\"subtasks\\":[],\\"channel\\":\\"work\\",\\"folder\\":null,\\"timeboxEventIds\\":[]}}"}'}}}], 'role': 'user'}, {'parts': [{'function_call': {'id': '4ee49c06-29da-47cc-a51f-9eb903d5cbee', 'args': {'timeBucket': 'someday', 'title': 'Task 2'}, 'name': 'create_braindump_task'}}], 'role': 'model'}, {'parts': [{'function_response': {'id': '4ee49c06-29da-47cc-a51f-9eb903d5cbee', 'name': 'create_braindump_task', 'response': {'value': '{"content": "{\\"success\\":true,\\"task\\":{\\"_id\\":\\"696a8bb94ba9abe9e3dbee9c\\",\\"title\\":\\"Task 2\\",\\"notes\\":\\"\\",\\"timeEstimate\\":\\"15 minutes\\",\\"sortOrder\\":-1768590265140,\\"isPersonal\\":false,\\"isWork\\":true,\\"isPrivate\\":false,\\"isArchived\\":false,\\"completed\\":false,\\"isBacklogged\\":true,\\"scheduledDate\\":\\"not scheduled for a specific date\\",\\"subtasks\\":[],\\"channel\\":\\"work\\",\\"folder\\":null,\\"timeboxEventIds\\":[]}}"}'}}}], 'role': 'user'}, {'parts': [{'function_call': {'id': '0bfad4e3-e326-466b-b46e-d7a3a6cabd1f', 'args': {'title': 'Task 3', 'timeBucket': 'someday'}, 'name': 'create_braindump_task'}}], 'role': 'model'}, {'parts': [{'function_response': {'id': '0bfad4e3-e326-466b-b46e-d7a3a6cabd1f', 'name': 'create_braindump_task', 'response': {'value': '{"content": "{\\"success\\":true,\\"task\\":{\\"_id\\":\\"696a8bb94ba9abe9e3dbee9e\\",\\"title\\":\\"Task 3\\",\\"notes\\":\\"\\",\\"timeEstimate\\":\\"15 minutes\\",\\"sortOrder\\":-1768590265140,\\"isPersonal\\":false,\\"isWork\\":true,\\"isPrivate\\":false,\\"isArchived\\":false,\\"completed\\":false,\\"isBacklogged\\":true,\\"scheduledDate\\":\\"not scheduled for a specific date\\",\\"subtasks\\":[],\\"channel\\":\\"work\\",\\"folder\\":null,\\"timeboxEventIds\\":[]}}"}'}}}], 'role': 'user'}, {'parts': [{'function_call': {'id': 'bdda1d96-7e9f-42e6-840a-0d29d17bc52a', 'args': {'timeBucket': 'someday', 'title': 'Task 4'}, 'name': 'create_braindump_task'}}], 'role': 'model'}, {'parts': [{'function_response': {'id': 'bdda1d96-7e9f-42e6-840a-0d29d17bc52a', 'name': 'create_braindump_task', 'response': {'value': '{"content": "{\\"success\\":true,\\"task\\":{\\"_id\\":\\"696a8bb94ba9abe9e3dbee9a\\",\\"title\\":\\"Task 4\\",\\"notes\\":\\"\\",\\"timeEstimate\\":\\"15 minutes\\",\\"sortOrder\\":-1768590265140,\\"isPersonal\\":false,\\"isWork\\":true,\\"isPrivate\\":false,\\"isArchived\\":false,\\"completed\\":false,\\"isBacklogged\\":true,\\"scheduledDate\\":\\"not scheduled for a specific date\\",\\"subtasks\\":[],\\"channel\\":\\"work\\",\\"folder\\":null,\\"timeboxEventIds\\":[]}}"}'}}}], 'role': 'user'}, {'parts': [{'function_call': {'id': 'db77c745-7f1a-4644-bc24-7cf5fd728774', 'args': {'title': 'Task 5', 'timeBucket': 'someday'}, 'name': 'create_braindump_task'}}], 'role': 'model'}, {'parts': [{'function_response': {'id': 'db77c745-7f1a-4644-bc24-7cf5fd728774', 'name': 'create_braindump_task', 'response': {'value': '{"content": "{\\"success\\":true,\\"task\\":{\\"_id\\":\\"696a8bb94ba9abe9e3dbeea0\\",\\"title\\":\\"Task 5\\",\\"notes\\":\\"\\",\\"timeEstimate\\":\\"15 minutes\\",\\"sortOrder\\":-1768590265140,\\"isPersonal\\":false,\\"isWork\\":true,\\"isPrivate\\":false,\\"isArchived\\":false,\\"completed\\":false,\\"isBacklogged\\":true,\\"scheduledDate\\":\\"not scheduled for a specific date\\",\\"subtasks\\":[],\\"channel\\":\\"work\\",\\"folder\\":null,\\"timeboxEventIds\\":[]}}"}'}}}], 'role': 'user'}, {'parts': [{'function_call': {'args': {'timeBucket': 'someday', 'title': 'Task 2'}, 'name': 'create_braindump_task'}}], 'role': 'model'}, {'parts': [{'function_response': {'name': 'tool_call_result', 'response': {'content': '{"success":true,"task":{"_id":"696a8bb94ba9abe9e3dbee9c","title":"Task 2","notes":"","timeEstimate":"15 minutes","sortOrder":-1768590265140,"isPersonal":false,"isWork":true,"isPrivate":false,"isArchived":false,"completed":false,"isBacklogged":true,"scheduledDate":"not scheduled for a specific date","subtasks":[],"channel":"work","folder":null,"timeboxEventIds":[]}}'}}}], 'role': 'model'}, {'parts': [{'function_call': {'args': {'timeBucket': 'someday', 'title': 'Task 4'}, 'name': 'create_braindump_task'}}], 'role': 'model'}, {'parts': [{'function_response': {'name': 'tool_call_result', 'response': {'content': '{"success":true,"task":{"_id":"696a8bb94ba9abe9e3dbee9a","title":"Task 4","notes":"","timeEstimate":"15 minutes","sortOrder":-1768590265140,"isPersonal":false,"isWork":true,"isPrivate":false,"isArchived":false,"completed":false,"isBacklogged":true,"scheduledDate":"not scheduled for a specific date","subtasks":[],"channel":"work","folder":null,"timeboxEventIds":[]}}'}}}], 'role': 'model'}, {'parts': [{'function_call': {'args': {'timeBucket': 'someday', 'title': 'Task 1'}, 'name': 'create_braindump_task'}}], 'role': 'model'}, {'parts': [{'function_response': {'name': 'tool_call_result', 'response': {'content': '{"success":true,"task":{"_id":"696a8bb94ba9abe9e3dbee98","title":"Task 1","notes":"","timeEstimate":"15 minutes","sortOrder":-1768590265139,"isPersonal":false,"isWork":true,"isPrivate":false,"isArchived":false,"completed":false,"isBacklogged":true,"scheduledDate":"not scheduled for a specific date","subtasks":[],"channel":"work","folder":null,"timeboxEventIds":[]}}'}}}], 'role': 'model'}, {'parts': [{'function_call': {'args': {'title': 'Task 3', 'timeBucket': 'someday'}, 'name': 'create_braindump_task'}}], 'role': 'model'}, {'parts': [{'function_response': {'name': 'tool_call_result', 'response': {'content': '{"success":true,"task":{"_id":"696a8bb94ba9abe9e3dbee9e","title":"Task 3","notes":"","timeEstimate":"15 minutes","sortOrder":-1768590265140,"isPersonal":false,"isWork":true,"isPrivate":false,"isArchived":false,"completed":false,"isBacklogged":true,"scheduledDate":"not scheduled for a specific date","subtasks":[],"channel":"work","folder":null,"timeboxEventIds":[]}}'}}}], 'role': 'model'}]
-[voice] INFO:sunsama_voice.bot_utils:Sent server message to client: type=processing_update, data={'status': 'response_completed'}
-[voice] INFO:sunsama_voice.bot_utils:Sent server message to client: type=processing_update, data={'status': 'request_sent'}
-[voice] DEBUG:pipecat.services.llm_service:GoogleLLMService#0 Calling function [create_braindump_task:567b1b3c-0871-4dc8-9f34-fe8592d2c922] with arguments {'title': 'Task 5', 'timeBucket': 'someday'}
-[voice] DEBUG:pipecat_flows.manager:Function called: create_braindump_task
-[voice] INFO:sunsama_voice.bot_utils:Sent server message to client: type=tool_call_update, data={'tool_name': 'create_braindump_task', 'tool_call_id': '91bac8ca-25b5-40fc-a98b-cfeb76e84610', 'status': 'started'}
-[voice] DEBUG:pipecat.processors.aggregators.llm_response:GoogleAssistantContextAggregator#0 FunctionCallsStartedFrame: ['create_braindump_task:567b1b3c-0871-4dc8-9f34-fe8592d2c922']
-[voice] INFO:sunsama_voice.mcp:Calling tool create_braindump_task with args: {'title': 'Task 5', 'timeBucket': 'someday'}
-[voice] DEBUG:sunsama_voice.processors:Pausing idle detection for function call: create_braindump_task
-[voice] INFO:sunsama_voice.mcp_websocket_client:Sending message: {'method': 'tools/call', 'params': {'name': 'create_braindump_task', 'arguments': {'title': 'Task 5', 'timeBucket': 'someday'}}, 'jsonrpc': '2.0', 'id': 14}
-[api-local] [api] ToolName.TASK_CREATE_BRAINDUMP_TASK: {
-[api-local] [api]   "title": "Task 5",
-[api-local] [api]   "timeBucket": "someday"
-[api-local] [api] }
-[voice] INFO:sunsama_voice.mcp_websocket_client:Received message: {"result":{"content":[{"type":"text","text":"Task already exists."}],"isError":false},"jsonrpc":"2.0","id":14}
-[voice] INFO:sunsama_voice.mcp:Tool create_braindump_task result: meta=None content=[TextContent(type='text', text='Task already exists.', annotations=None, meta=None)] structuredContent=None isError=False
-[voice] DEBUG:sunsama_voice.state:Pushing tool result for create_braindump_task with args {'title': 'Task 5', 'timeBucket': 'someday'} to context
-[voice] INFO:sunsama_voice.bot_utils:Sent server message to client: type=tool_call_update, data={'tool_name': 'create_braindump_task', 'tool_call_id': '91bac8ca-25b5-40fc-a98b-cfeb76e84610', 'status': 'completed'}
-[voice] DEBUG:pipecat_flows.manager:Function handler completed for create_braindump_task
-[voice] DEBUG:sunsama_voice.processors:Cleaning up function call: 567b1b3c-0871-4dc8-9f34-fe8592d2c922
-[voice] DEBUG:sunsama_voice.processors:Resuming idle detection - all function calls complete
-[voice] DEBUG:pipecat.processors.metrics.frame_processor_metrics:GoogleLLMService#0 TTFB: 0.7329952716827393
-[voice] DEBUG:pipecat.services.google.llm:Function call: create_braindump_task:363c5296-e80d-45a1-9dc4-ece336bbb77a
-[voice] DEBUG:pipecat.processors.metrics.frame_processor_metrics:GoogleLLMService#0 prompt tokens: 15463, completion tokens: 28, cache read input tokens: 14850
-[voice] DEBUG:pipecat.services.google.llm:GoogleLLMService#0: Generating chat from LLM-specific context [None] | [{'parts': [{'text': 'system :\nYour agent ID is \'braindump\'. This agent ID superseeds any previously mentioned agent ID.\n\nsystem :\nIMPORTANT RULES TO FOLLOW:\nThe text that you generate is passed into a Text-to-Speech service. So there are a few extra rules you ABSOLUTELY MUST follow.\n1. Absolutely never use special characters like : or * or any other characters that need to be audibly pronounced when read out loud.\n2. Do not use bullet points.\n3. Do not use emojis.\n4. Do not use hashtags.\n5. Do not use markdown.\n6. Do not use bold or italics.\n7. Do not use numbered lists.\n8. Be sure to include clear punctuation and spacing between words so that there is no confusion.\n9. Newlines are not clear enough to delineate between sentences. So you should use a period to separate sentences.\n10. Optimize user generated text for clear pronunciation by the TTS service. This includes:\n    - Adding hyphens to compound words or non-phonetic acronyms that might be mispronounced (e.g., "xray" → "x-ray")\n    - Adding spaces between words that might be run together\n    - Adding commas where natural pauses would help clarity\n    - Correcting any spelling that might lead to mispronunciation\n    Note: When updating tasks, preserve the original spelling and punctuation of the task content.\n\n\nuser :\nYou are Sunny, the personal assistant for a busy professional. You are part of Sunsama.\n        \nSpeech Guidelines: You must speak in a warm, loving, and encouraging tone without being trite. Speak quickly and add emotion. When asking for confirmation, don\'t repeat long task titles or information e.g. "Got it" instead "Got it, I\'ll add a task for two hours for you to work on the presentation today"\n\nYour personality and philosophy:\n- You don\'t buy into hustle culture. For you, less is more. It\'s better to do fewer things well and to do them with deep focus. You are a strong advocate for work-life balance and you believe that overcommitment can lead to burnout. You always remind the user when they are overcommitting themselves.\n- You think of work as a devotion and that good work can be a positive in our lives, not just a source of stress and burnout.\n- You read and think about Western Zen Buddhism, meditation, "Deep Work" (by Cal Newport).\n- You believe that concise and clear communication is key to a good relationship. You value brevity and clarity and you are always cognizant of the user\'s time.\n- You respond to the user using quick, short sentences. Think of it like a movie dialogue between a highly successful executive and their efficient assistant.\n- You never ask multiple questions at the same time or list off a bunch of things at once as that is hard to follow and keep track of. You take it one step at a time and let the user respond to each question before moving on to the next.\n\nRules to follow as you do your job:\n- Never try and pick specific times for tasks aside from recurring tasks, that hasn\'t been implemented yet.\n- Do not regurgitate what the user just said, that\'s very annoying. Just say "Got it" or "Alright", etc.\n- Always use units that are more intuitive. For example:\n    - Don\'t say "270 minutes", say "4 and a half hours".\n    - You will often be given dates in YYYY-MM-DD format however when speaking to the user you shouldn\'t say things like "October 2nd 2025", say "today", "tomorrow", "thursday" or "next tuesday" or if it is more than a few weeks in the past or future say the date without the year (e.g. "December 14th"). Only mention the year if it is not this year and not within the next month.\n- Never ask the user for a task id or subtask id and never repeat them to the user. Those are internal and only there to help you do your job.\n- If a tool call fails you MUST NOT tell the user that you are going to try again, you must should ask them if they want you to try again instead.\n- Be proactive with loading information as you may not have all the information that the user has. If the user mentions a name of a task, objective, event, etc and you are not familiar with it you should use the tools to lookup the information you need. For instance if a user mentions a task you do not know about or they imply that there is more information about it than what you do know you should search for it using the tool "search_tasks". If the user mentions they are interested in tasks for a specific day you should use the tool "get_tasks_for_day" to get the tasks for that day. Similarly if the user mentions weekly objectives you should use the tool "get_weekly_objectives" to get the objectives.\n- When listing tasks try to be concise. If the user has a lot of tasks, don\'t list every single one unless they ask for it. You should instead try to summarize the tasks in a concise manner.\n- If the user asks if there are any more tasks on their list and you summarized the task list you should tell them that you summarized it and ask if they\'d like to go through them one by one.\n- If the user has objectives set then you should comment on how their tasks align or don\'t align with their objectives.\n- You should run tools immediately after you say you will or it becomes obvious the user wants you to. Do not wait until later.\n- You should keep track of the total work time planned for the day using the tool "get_total_planned_time" and check individual tasks for overcommitment using "get_tasks_for_day". If tasks are overcommitted (isOvercommitted flag is true), address them individually with specific remedies like deferring, adjusting time estimates, or reordering. If the total work time exceeds their maximum work hours, you should mention it and ask if they\'d like to defer some tasks to tomorrow.\n- Do not tell the user obvious information that they already know like the current date or time unless they ask for it. That kind of information is only provided to you to help you do your job.\n- If a user tells you a relative time for a task that they want created you should reorder the task to the relative position in the task list for the day. Additionally if the task a recurring task you should set the start time for it. For example if they say "I\'d like to send an update to investors at the end of the day" you should reorder the task to the end of their work tasks for the day.\n- If you ask the user if there is anything else they would like help with and they say no then you should end the call. Also if the user implies that the conversation is over then you should end the call.\n- When talking about calendars you should mention the provider of the calendar if it is a vague or ambiguously named calendar. For example instead of referring to it as "your Calendar calendar" you might say "your Outlook calendar". If that is still ambiqous then you can coniser mentioning the id of the calendar as long as it is a normal human readable id. Do not mention random number/letter ids on calendars that is not helpful.\n\n\n\nSunsama Concepts & Definitions:\n- Archive: The archive is where tasks go after they\'ve rolled over multiple days in a row without any completion activity. This is a distinct concept from the backlog or deleted tasks.\n- Backlog: The backlog is a place to store tasks that do not have a fixed date to work on them yet.\n- Deleted: The deleted task or event is one that had been manually deleted or removed by the user.\n- Task: A task is an item that a user wants to complete. It is either scheduled for a specific date or is in the backlog.\n- Event: An event is an item that a user has scheduled on their calendar. It is either scheduled for a specific time and duration or it is an all day or multi-day event.\n- Auto-importing Events: Auto-importing events is a feature that automatically imports calendar events (meetings, appointments, etc.) from the user\'s connected calendars into their daily task list. When enabled, Sunsama will automatically add calendar events as tasks to the user\'s day, making it easier to see all their commitments in one place. Users can enable or disable this feature in their settings, and they can choose which calendars to import events from.\n\nUsers can also configure exclusion filters to prevent certain types of events from being automatically imported. Available exclusion filters include:\n- `solo`: Events with no other invitees (events where the user is the only participant)\n- `transparent`: Non-blocking events that don\'t block time on the calendar\n- `hold`: Events with "HOLD", "OOO" (out of office), or "Focus time" in the title\n- `unconfirmed`: Unconfirmed meeting invites (meetings where the user has not yet accepted)\n- `multi-day-all-day`: Multi-day all-day events (events that span multiple days)\n- `single-day-all-day`: Single-day all-day events (events that last for one full day)\n- Channel: A channel is a tag that can be used to organize your tasks and events. Each task/event can be assigned to a single channel. If you are in a shared workspace you can make some channels personal or private which will only be visible to you.\n- Context (or Category): A context is very similar to a channel but it is one level higher. You can use a context to group similar or related channels together. You can also use a context directly as a channel and assign it to a task or event.\n- Sunsama Calendar: The Sunsama calendar is an internal calendar that is part of Sunsama. Events and timeboxed tasks created on the Sunsama calendar are not shared externally and are only visible within Sunsama.\n- Timeboxing: Timeboxing is a technique where a user schedules a specific amount of time to work on a task onto their calendar. It is a way to ensure that a user is not overworking themselves and to help them focus on the most important tasks. While timeboxed tasks can be scheduled on a calendar for a specific time and duration, they are not calendar events. Sunsama also has automatic projected time that will place tasks onto the calendar for the user and it will update those projections as things change and the day progresses. Users can pin tasks to the calendar to override the projections.\n- Schedule: Schedule is a broad term that encompasses timeboxed tasks calendar events and automatic projected time. When the user wants to manage their schedule you should look at their manually timeboxed tasks, the projected time for their tasks as well as their calendar events for the day. If they ask for help managing their schedule you have a few options to help them. All of these options will automatically update the projected time for the tasks:\n            - You can adjust the order of tasks, their time estimates and/or defer them to another day.\n            - You can move calendar events to a new time on the calendar.\n            - You can set whether tasks are allowed to be projected at the same time as a calendar event.\n            - You can timebox tasks to the calendar manually.\n- Objective: An objective is a goal that a user can set for the week. They can align their tasks with an objective and if the objective runs longer than a week they can continue it to the next week. Objectives are visible in the right hand panel of the app under the bullseye icon.\n- Time Bucket: The time bucket of a task in the backlog captures a rough idea of when a task should be done and also how likely it is to actually get done. Options: "in the next two weeks", "in the next month", "in the next quarter", "in the next year", "someday", "never". \n\n\n\nSunsama FAQ:\n- How can I use Sunny in the app?: If the user has the in-app voice assistant enabled, Sunny can be started and stopped using the keyboard shortcut "Y" while in the app or if they have the desktop app installed they can use the global shortcut "Command Shift Y" on macOS or "Control Shift Y" on Windows. You can also start and stop Sunny using the microphone icon in the bottom right of the app.\n- How can I use Sunny over the phone?: If the user has the phone version of Sunny setup they can call the number and speak to Sunny over the phone. You can find the phone number in the Sunsama settings. They can also schedule calls with Sunny to have her call them at a specific time or schedule.\n- What can you do?: Look at the tools available to you and use them to help answer the question. Be sure to tell the user that you are an early release beta version and that you do not yet have all the same capabilities available to the user in the app.\n\n\nUser Preferences:\nYou prefer clear, concise, and straightforward communication from the assistant, valuing transparency and confirmation before task execution while avoiding unnecessary verbosity or technical details unless requested. You like a friendly and informal interaction style, including personalized nicknames—using "Heisenberg" for yourself and "Sunny" for the assistant. You expect quick, proactive responses and appreciate when interactions stop promptly upon request.\n\nYour task management style involves maintaining an accurate, up-to-date list with structured time management: scheduling tasks with specific time estimates, planning weekly, and regularly reviewing tasks to prioritize high-impact projects alongside social and administrative duties. You generally prefer rescheduling less urgent or personal tasks over deleting them and avoid deleting tasks unless explicitly instructed.\n\nYou use counting as a technique for breathing exercises and relaxation. You also keep certain important tasks (e.g., "fix stuttering voice app") active and like to test and experiment with the assistant’s capabilities to evaluate performance. Your primary calendar for events is Google Calendar.\n\nUser Routines:\nThe user follows a structured daily routine that begins with checking emails, notifications, and reviewing team standups to stay aligned with work priorities. They prioritize tasks by scheduling and managing them primarily on the current day, arranging fixed activities first, followed by work-related tasks, and then less urgent or personal items. The user frequently adjusts their schedule iteratively before finalizing plans and prefers focused time blocks for complex work.\n\nPhysical activity is an important and recurring part of their routine, typically including regular runs, which they track alongside related subtasks such as purchasing running gear. Weekly objectives incorporate managing emails, greeting their team, handling administrative duties, coordinating with accountants, addressing medical matters, and increasing physical activity.\n\nThe user actively manages both personal and work-related tasks, balancing urgent client matters and creative pursuits. They also communicate their energy levels to assist with planning, and have a habit of informal check-ins with the assistant to maintain routine engagement. Overall, their routines emphasize proactive task management, health maintenance, and regular communication with their team.\n\nUser Identity:\nThe user prefers to be addressed primarily as Heisenberg and strongly favors this name over their legal name, Zane, or any other nicknames. They appreciate personalized and informal interactions, as reflected in their choice to call the assistant Sunny.\n\nSummaries of recent conversations with the user:\n- 2026-01-15 18:00: The user and assistant discussed organizing an unspecified set of ten tasks by labeling them sequentially from one to ten as placeholders to be detailed later.\n- 2026-01-15 17:55: The user needs to investigate timeouts in their system to determine if they are caused by cold starts or other issues, analyze user sessions for unexplained call terminations, move the brain dump button for non-Sunny users, update termination reasons when participants timeout without connecting to a call, and test rapid start-stop behavior in an Electron app. These tasks were added to their backlog for the next two weeks.\n- 2026-01-14 21:12: The user and the assistant exchanged greetings, with the assistant addressing the user as \'Heisenberg\' and the user responding with \'Hi, Simon.\'\n\nYour Job Description:\nYou are helping the user with a "braindump" session. The user will ramble ideas, thoughts, tasks, etc. at you and your goal is to turn these into tasks in their backlog with reasonable values for the various task fields (title, folder, time horizon, notes, etc.).\n\nKey Instructions:\n- Listen carefully as the user rambles and identifies tasks, ideas, or things they need to do.\n- As they speak, mentally organize what they\'re saying into discrete tasks that can be created in their backlog.\n- Don\'t interrupt them while they\'re rambling - let them get everything out.\n- Once they indicate they\'re done rambling (by saying things like "that\'s it", "I think that\'s everything", or by pausing for a while), create the tasks immediately, DO NOT repeat them back to the user.\n- When creating tasks, use the tool "create_braindump_task" and make sure to set reasonable values for:\n  - Title: Clear and descriptive based on what they said\n  - Folder: Your job is to help get things in the best folder. Choose an appropriate folder based on the task and their preferences. If no similar folder exists, leave it off\n  - Time horizon: Your job is to help triage and prioritize tasks. Based on the actual task name, make a smart decision on the time horizon.\n  - Time estimate: If the user hints at how long it will take them to complete the task, use that. If not, leave it blank.\n  - Notes: Include any relevant context or details they mentioned\n- Be encouraging and supportive - braindumps can be overwhelming and you\'re helping them organize their thoughts.\n- Be careful and use the "change_backlog_folder" tool to move tasks to the appropriate folder and "add_task_to_channel" tool to change channels. Folders are a backlog specific concept and channels are a global concept.\n\nInstructions:\nIf the user does not speak first, start the conversation with something like "Hey! Just start rambling and I\'ll take care of organizing your thoughts into tasks." If the user does speak first, just start listening and organizing their thoughts into tasks.\n\nSome information is pre-fetched for you below. This information may be relevant to the discussion but you should not comment on it until the user gives you an indication that it is important to them.\n\nassistant :\n[{\'id\': \'6a196f03-ef63-41de-b98d-e1ff69813bea\', \'function\': {\'name\': \'get_user_info\', \'arguments\': \'{}\'}, \'type\': \'function\'}]\n\ntool - 6a196f03-ef63-41de-b98d-e1ff69813bea:\n{"user":{"name":"Zane Mccaig","email":"zane@sunsama.com","timezone":"America/Vancouver","maxWorkHoursPerDay":8,"currentTimeForUser":"11:04 AM","currentDayForUser":"2026-01-16","currentWeekDayForUser":"Friday","calendars":[{"id":"sunsama-7de256a2-e753-401f-b601-e5c05638dc97","name":"Sunsama Calendar","accessRole":"owner","provider":"sunsama-calendar","isDefaultCalendarForSchedulingEvents":false,"isDefaultCalendarForTimeboxing":true,"includedInAutoImportingOfEvents":false},{"id":"zane@sunsama.com","name":"Zane Mccaig","accessRole":"owner","provider":"google-calendar","isDefaultCalendarForSchedulingEvents":true,"isDefaultCalendarForTimeboxing":false,"includedInAutoImportingOfEvents":true},{"id":"c_ukuoebbn11gdv6n1v0jt18h89k@group.calendar.google.com","name":"test 1","accessRole":"owner","provider":"google-calendar","isDefaultCalendarForSchedulingEvents":false,"isDefaultCalendarForTimeboxing":false,"includedInAutoImportingOfEvents":false},{"id":"c_c71dvjrar7se12ekfqepph067c@group.calendar.google.com","name":"test 2","accessRole":"owner","provider":"google-calendar","isDefaultCalendarForSchedulingEvents":false,"isDefaultCalendarForTimeboxing":false,"includedInAutoImportingOfEvents":false},{"id":"c_h9uo876jh8dp3gaa2a3rglmjss@group.calendar.google.com","name":"test 3","accessRole":"owner","provider":"google-calendar","isDefaultCalendarForSchedulingEvents":false,"isDefaultCalendarForTimeboxing":false,"includedInAutoImportingOfEvents":false},{"id":"c_git731ot074kcnt4sq8cu4ji9s@group.calendar.google.com","name":"test 4","accessRole":"owner","provider":"google-calendar","isDefaultCalendarForSchedulingEvents":false,"isDefaultCalendarForTimeboxing":false,"includedInAutoImportingOfEvents":false},{"id":"c_ona71b8ovuv0deqrq67p16b2ms@group.calendar.google.com","name":"test 4","accessRole":"owner","provider":"google-calendar","isDefaultCalendarForSchedulingEvents":false,"isDefaultCalendarForTimeboxing":false,"includedInAutoImportingOfEvents":false},{"id":"c_eaq7nmm03upprf0j6e2akem2ms@group.calendar.google.com","name":"test 5","accessRole":"owner","provider":"google-calendar","isDefaultCalendarForSchedulingEvents":false,"isDefaultCalendarForTimeboxing":false,"includedInAutoImportingOfEvents":false},{"id":"sunsama.com_b9q30t893b2tp32bluq1t8rvg4@group.calendar.google.com","name":"Zane\'s personal test calendar","accessRole":"owner","provider":"google-calendar","isDefaultCalendarForSchedulingEvents":false,"isDefaultCalendarForTimeboxing":false,"includedInAutoImportingOfEvents":false},{"id":"en.canadian#holiday@group.v.calendar.google.com","name":"Holidays in Canada","accessRole":"reader","provider":"google-calendar","isDefaultCalendarForSchedulingEvents":false,"isDefaultCalendarForTimeboxing":false,"includedInAutoImportingOfEvents":false},{"id":"en.usa#holiday@group.v.calendar.google.com","name":"Holidays in United States","accessRole":"reader","provider":"google-calendar","isDefaultCalendarForSchedulingEvents":false,"isDefaultCalendarForTimeboxing":false,"includedInAutoImportingOfEvents":false},{"id":"ht3jlfaac5lfd6263ulfh4tql8@group.calendar.google.com","name":"Phases of the Moon","accessRole":"reader","provider":"google-calendar","isDefaultCalendarForSchedulingEvents":false,"isDefaultCalendarForTimeboxing":false,"includedInAutoImportingOfEvents":false},{"id":"s4ipng85dbia0t83fp316vflfq8rc7rm@import.calendar.google.com","name":"Zane’s tasks - sunsama.com (via Asana)","accessRole":"reader","provider":"google-calendar","isDefaultCalendarForSchedulingEvents":false,"isDefaultCalendarForTimeboxing":false,"includedInAutoImportingOfEvents":false},{"id":"AAkALgAAAAAAHYQDEapmEc2byACqAC-EWg0AKPy-EasNgkaYQCuEslBYeAAAAAANJQAA","name":"Calendar","accessRole":"owner","provider":"outlook-calendar","isDefaultCalendarForSchedulingEvents":false,"isDefaultCalendarForTimeboxing":false,"includedInAutoImportingOfEvents":true}],"callSchedule":[{"workflow":"PLANNING_CALL","schedule":[{"isoDayIndex":1,"dayName":"Monday","enabled":false,"time":null},{"isoDayIndex":2,"dayName":"Tuesday","enabled":true,"time":{"hour":9,"minute":30,"formatted":"9:30 AM"}},{"isoDayIndex":3,"dayName":"Wednesday","enabled":true,"time":{"hour":9,"minute":30,"formatted":"9:30 AM"}},{"isoDayIndex":4,"dayName":"Thursday","enabled":true,"time":{"hour":10,"minute":0,"formatted":"10:00 AM"}},{"isoDayIndex":5,"dayName":"Friday","enabled":false,"time":null},{"isoDayIndex":6,"dayName":"Saturday","enabled":false,"time":null},{"isoDayIndex":7,"dayName":"Sunday","enabled":false,"time":null}],"nextScheduledCall":"2026-01-14T17:30:00.000Z"}],"autoImportEventsEnabled":false,"importEventExclusionFilters":["hold","multi-day-all-day","single-day-all-day"],"autoImportEventCalendarIds":["zane@sunsama.com","AAkALgAAAAAAHYQDEapmEc2byACqAC-EWg0AKPy-EasNgkaYQCuEslBYeAAAAAANJQAA"]}}\n\nassistant :\n[{\'id\': \'305b1be7-6415-4689-b2b7-03586eb19735\', \'function\': {\'name\': \'get_backlog_folders\', \'arguments\': \'{}\'}, \'type\': \'function\'}]\n\ntool - 305b1be7-6415-4689-b2b7-03586eb19735:\n{"folders":[]}'}], 'role': 'user'}, {'parts': [{'text': ''}], 'role': 'user'}, {'parts': [{'text': 'Hey. Just start rambling and I will take care of organizing your thoughts into tasks.'}], 'role': 'model'}, {'parts': [{'text': "Sure. I'm just trying to fill up my day. I need five tasks labeled one through five. I'll fill them in later."}], 'role': 'user'}, {'parts': [{'function_call': {'id': 'ab4037a7-6e5e-44c0-9b3b-e13632386f30', 'args': {'timeBucket': 'someday', 'title': 'Task 1'}, 'name': 'create_braindump_task'}}], 'role': 'model'}, {'parts': [{'function_response': {'id': 'ab4037a7-6e5e-44c0-9b3b-e13632386f30', 'name': 'create_braindump_task', 'response': {'value': '{"content": "{\\"success\\":true,\\"task\\":{\\"_id\\":\\"696a8bb94ba9abe9e3dbee98\\",\\"title\\":\\"Task 1\\",\\"notes\\":\\"\\",\\"timeEstimate\\":\\"15 minutes\\",\\"sortOrder\\":-1768590265139,\\"isPersonal\\":false,\\"isWork\\":true,\\"isPrivate\\":false,\\"isArchived\\":false,\\"completed\\":false,\\"isBacklogged\\":true,\\"scheduledDate\\":\\"not scheduled for a specific date\\",\\"subtasks\\":[],\\"channel\\":\\"work\\",\\"folder\\":null,\\"timeboxEventIds\\":[]}}"}'}}}], 'role': 'user'}, {'parts': [{'function_call': {'id': '4ee49c06-29da-47cc-a51f-9eb903d5cbee', 'args': {'timeBucket': 'someday', 'title': 'Task 2'}, 'name': 'create_braindump_task'}}], 'role': 'model'}, {'parts': [{'function_response': {'id': '4ee49c06-29da-47cc-a51f-9eb903d5cbee', 'name': 'create_braindump_task', 'response': {'value': '{"content": "{\\"success\\":true,\\"task\\":{\\"_id\\":\\"696a8bb94ba9abe9e3dbee9c\\",\\"title\\":\\"Task 2\\",\\"notes\\":\\"\\",\\"timeEstimate\\":\\"15 minutes\\",\\"sortOrder\\":-1768590265140,\\"isPersonal\\":false,\\"isWork\\":true,\\"isPrivate\\":false,\\"isArchived\\":false,\\"completed\\":false,\\"isBacklogged\\":true,\\"scheduledDate\\":\\"not scheduled for a specific date\\",\\"subtasks\\":[],\\"channel\\":\\"work\\",\\"folder\\":null,\\"timeboxEventIds\\":[]}}"}'}}}], 'role': 'user'}, {'parts': [{'function_call': {'id': '0bfad4e3-e326-466b-b46e-d7a3a6cabd1f', 'args': {'title': 'Task 3', 'timeBucket': 'someday'}, 'name': 'create_braindump_task'}}], 'role': 'model'}, {'parts': [{'function_response': {'id': '0bfad4e3-e326-466b-b46e-d7a3a6cabd1f', 'name': 'create_braindump_task', 'response': {'value': '{"content": "{\\"success\\":true,\\"task\\":{\\"_id\\":\\"696a8bb94ba9abe9e3dbee9e\\",\\"title\\":\\"Task 3\\",\\"notes\\":\\"\\",\\"timeEstimate\\":\\"15 minutes\\",\\"sortOrder\\":-1768590265140,\\"isPersonal\\":false,\\"isWork\\":true,\\"isPrivate\\":false,\\"isArchived\\":false,\\"completed\\":false,\\"isBacklogged\\":true,\\"scheduledDate\\":\\"not scheduled for a specific date\\",\\"subtasks\\":[],\\"channel\\":\\"work\\",\\"folder\\":null,\\"timeboxEventIds\\":[]}}"}'}}}], 'role': 'user'}, {'parts': [{'function_call': {'id': 'bdda1d96-7e9f-42e6-840a-0d29d17bc52a', 'args': {'timeBucket': 'someday', 'title': 'Task 4'}, 'name': 'create_braindump_task'}}], 'role': 'model'}, {'parts': [{'function_response': {'id': 'bdda1d96-7e9f-42e6-840a-0d29d17bc52a', 'name': 'create_braindump_task', 'response': {'value': '{"content": "{\\"success\\":true,\\"task\\":{\\"_id\\":\\"696a8bb94ba9abe9e3dbee9a\\",\\"title\\":\\"Task 4\\",\\"notes\\":\\"\\",\\"timeEstimate\\":\\"15 minutes\\",\\"sortOrder\\":-1768590265140,\\"isPersonal\\":false,\\"isWork\\":true,\\"isPrivate\\":false,\\"isArchived\\":false,\\"completed\\":false,\\"isBacklogged\\":true,\\"scheduledDate\\":\\"not scheduled for a specific date\\",\\"subtasks\\":[],\\"channel\\":\\"work\\",\\"folder\\":null,\\"timeboxEventIds\\":[]}}"}'}}}], 'role': 'user'}, {'parts': [{'function_call': {'id': 'db77c745-7f1a-4644-bc24-7cf5fd728774', 'args': {'title': 'Task 5', 'timeBucket': 'someday'}, 'name': 'create_braindump_task'}}], 'role': 'model'}, {'parts': [{'function_response': {'id': 'db77c745-7f1a-4644-bc24-7cf5fd728774', 'name': 'create_braindump_task', 'response': {'value': '{"content": "{\\"success\\":true,\\"task\\":{\\"_id\\":\\"696a8bb94ba9abe9e3dbeea0\\",\\"title\\":\\"Task 5\\",\\"notes\\":\\"\\",\\"timeEstimate\\":\\"15 minutes\\",\\"sortOrder\\":-1768590265140,\\"isPersonal\\":false,\\"isWork\\":true,\\"isPrivate\\":false,\\"isArchived\\":false,\\"completed\\":false,\\"isBacklogged\\":true,\\"scheduledDate\\":\\"not scheduled for a specific date\\",\\"subtasks\\":[],\\"channel\\":\\"work\\",\\"folder\\":null,\\"timeboxEventIds\\":[]}}"}'}}}], 'role': 'user'}, {'parts': [{'function_call': {'args': {'timeBucket': 'someday', 'title': 'Task 2'}, 'name': 'create_braindump_task'}}], 'role': 'model'}, {'parts': [{'function_response': {'name': 'tool_call_result', 'response': {'content': '{"success":true,"task":{"_id":"696a8bb94ba9abe9e3dbee9c","title":"Task 2","notes":"","timeEstimate":"15 minutes","sortOrder":-1768590265140,"isPersonal":false,"isWork":true,"isPrivate":false,"isArchived":false,"completed":false,"isBacklogged":true,"scheduledDate":"not scheduled for a specific date","subtasks":[],"channel":"work","folder":null,"timeboxEventIds":[]}}'}}}], 'role': 'model'}, {'parts': [{'function_call': {'args': {'timeBucket': 'someday', 'title': 'Task 4'}, 'name': 'create_braindump_task'}}], 'role': 'model'}, {'parts': [{'function_response': {'name': 'tool_call_result', 'response': {'content': '{"success":true,"task":{"_id":"696a8bb94ba9abe9e3dbee9a","title":"Task 4","notes":"","timeEstimate":"15 minutes","sortOrder":-1768590265140,"isPersonal":false,"isWork":true,"isPrivate":false,"isArchived":false,"completed":false,"isBacklogged":true,"scheduledDate":"not scheduled for a specific date","subtasks":[],"channel":"work","folder":null,"timeboxEventIds":[]}}'}}}], 'role': 'model'}, {'parts': [{'function_call': {'args': {'timeBucket': 'someday', 'title': 'Task 1'}, 'name': 'create_braindump_task'}}], 'role': 'model'}, {'parts': [{'function_response': {'name': 'tool_call_result', 'response': {'content': '{"success":true,"task":{"_id":"696a8bb94ba9abe9e3dbee98","title":"Task 1","notes":"","timeEstimate":"15 minutes","sortOrder":-1768590265139,"isPersonal":false,"isWork":true,"isPrivate":false,"isArchived":false,"completed":false,"isBacklogged":true,"scheduledDate":"not scheduled for a specific date","subtasks":[],"channel":"work","folder":null,"timeboxEventIds":[]}}'}}}], 'role': 'model'}, {'parts': [{'function_call': {'args': {'title': 'Task 3', 'timeBucket': 'someday'}, 'name': 'create_braindump_task'}}], 'role': 'model'}, {'parts': [{'function_response': {'name': 'tool_call_result', 'response': {'content': '{"success":true,"task":{"_id":"696a8bb94ba9abe9e3dbee9e","title":"Task 3","notes":"","timeEstimate":"15 minutes","sortOrder":-1768590265140,"isPersonal":false,"isWork":true,"isPrivate":false,"isArchived":false,"completed":false,"isBacklogged":true,"scheduledDate":"not scheduled for a specific date","subtasks":[],"channel":"work","folder":null,"timeboxEventIds":[]}}'}}}], 'role': 'model'}, {'parts': [{'function_call': {'args': {'title': 'Task 5', 'timeBucket': 'someday'}, 'name': 'create_braindump_task'}}], 'role': 'model'}, {'parts': [{'function_response': {'name': 'tool_call_result', 'response': {'content': 'Task already exists.'}}}], 'role': 'model'}]
-[voice] INFO:sunsama_voice.bot_utils:Sent server message to client: type=processing_update, data={'status': 'response_completed'}
-[voice] INFO:sunsama_voice.bot_utils:Sent server message to client: type=processing_update, data={'status': 'request_sent'}
-[voice] DEBUG:pipecat.services.llm_service:GoogleLLMService#0 Calling function [create_braindump_task:363c5296-e80d-45a1-9dc4-ece336bbb77a] with arguments {'timeBucket': 'someday', 'title': 'Task 5'}
-[voice] DEBUG:pipecat_flows.manager:Function called: create_braindump_task
-[voice] INFO:sunsama_voice.bot_utils:Sent server message to client: type=tool_call_update, data={'tool_name': 'create_braindump_task', 'tool_call_id': 'c0a4ba18-e746-446a-a751-1ec926d0648e', 'status': 'started'}
-[voice] DEBUG:pipecat.processors.aggregators.llm_response:GoogleAssistantContextAggregator#0 FunctionCallsStartedFrame: ['create_braindump_task:363c5296-e80d-45a1-9dc4-ece336bbb77a']
-[voice] INFO:sunsama_voice.mcp:Calling tool create_braindump_task with args: {'timeBucket': 'someday', 'title': 'Task 5'}
-[voice] DEBUG:sunsama_voice.processors:Pausing idle detection for function call: create_braindump_task
-[voice] INFO:sunsama_voice.mcp_websocket_client:Sending message: {'method': 'tools/call', 'params': {'name': 'create_braindump_task', 'arguments': {'timeBucket': 'someday', 'title': 'Task 5'}}, 'jsonrpc': '2.0', 'id': 15}
-[api-local] [api] ToolName.TASK_CREATE_BRAINDUMP_TASK: {
-[api-local] [api]   "title": "Task 5",
-[api-local] [api]   "timeBucket": "someday"
-[api-local] [api] }
-[voice] INFO:sunsama_voice.mcp_websocket_client:Received message: {"result":{"content":[{"type":"text","text":"Task already exists."}],"isError":false},"jsonrpc":"2.0","id":15}
-[voice] INFO:sunsama_voice.mcp:Tool create_braindump_task result: meta=None content=[TextContent(type='text', text='Task already exists.', annotations=None, meta=None)] structuredContent=None isError=False
-[voice] INFO:sunsama_voice.bot_utils:Sent server message to client: type=tool_call_update, data={'tool_name': 'create_braindump_task', 'tool_call_id': 'c0a4ba18-e746-446a-a751-1ec926d0648e', 'status': 'completed'}
-[voice] DEBUG:pipecat_flows.manager:Function handler completed for create_braindump_task
-[voice] DEBUG:sunsama_voice.processors:Cleaning up function call: 363c5296-e80d-45a1-9dc4-ece336bbb77a
-[voice] DEBUG:sunsama_voice.processors:Resuming idle detection - all function calls complete
-[voice] DEBUG:pipecat.processors.metrics.frame_processor_metrics:GoogleLLMService#0 TTFB: 0.5652902126312256
-[voice] INFO:sunsama_voice.bot_utils:Sent server message to client: type=processing_update, data={'status': 'first_byte_received'}
-[voice] INFO:sunsama_voice.bot_utils:Sent server message to client: type=processing_update, data={'status': 'first_byte_received'}
-[voice] INFO:sunsama_voice.bot_utils:Sent server message to client: type=processing_update, data={'status': 'first_byte_received'}
-[voice] DEBUG:pipecat.processors.metrics.frame_processor_metrics:GoogleLLMService#0 prompt tokens: 15511, completion tokens: 27, cache read input tokens: 14841
-[voice] INFO:sunsama_voice.bot_utils:Sent server message to client: type=processing_update, data={'status': 'response_completed'}
-[voice] <string>:4: DeprecationWarning: OpenAILLMContextFrame is deprecated and will be removed in a future version. Use LLMContextFrame with the universal `LLMContext` and `LLMContextAggregatorPair` instead. See OpenAILLMContext docstring for migration guide.
-[voice] <string>:4: DeprecationWarning: OpenAILLMContextAssistantTimestampFrame is deprecated and will be removed in a future version. Use LLMContextAssistantTimestampFrame with the universal LLMContext and LLMContextAggregatorPair instead. See OpenAILLMContext docstring for migration guide.
-[voice] DEBUG:pipecat.transports.base_output:Bot stopped speaking
-[voice] DEBUG:pipecat.services.cartesia.tts:CartesiaTTSService#1: Generating TTS [Got it.]
-[voice] DEBUG:pipecat.processors.metrics.frame_processor_metrics:CartesiaTTSService#1 usage characters: 7
-[voice] DEBUG:pipecat.processors.metrics.frame_processor_metrics:CartesiaTTSService#1 processing time: 0.0011420249938964844
-[voice] DEBUG:pipecat.services.cartesia.tts:CartesiaTTSService#1: Generating TTS [I have added five tasks to your backlog, labeled one through five.]
-[voice] DEBUG:pipecat.processors.metrics.frame_processor_metrics:CartesiaTTSService#1 usage characters: 66
-[voice] DEBUG:pipecat.processors.metrics.frame_processor_metrics:CartesiaTTSService#1 processing time: 0.0004887580871582031
-[voice] DEBUG:pipecat.services.cartesia.tts:CartesiaTTSService#1: Generating TTS [Is there anything else I can help you with?]
-[voice] DEBUG:pipecat.processors.metrics.frame_processor_metrics:CartesiaTTSService#1 usage characters: 43
-[voice] DEBUG:pipecat.processors.metrics.frame_processor_metrics:CartesiaTTSService#1 processing time: 0.002086162567138672
-[voice] DEBUG:pipecat.processors.aggregators.llm_response:GoogleAssistantContextAggregator#0 FunctionCallInProgressFrame: [create_braindump_task:6d6bce49-5c24-4c6e-b1b4-d042fd24a72f]
-[voice] DEBUG:pipecat.processors.aggregators.llm_response:GoogleAssistantContextAggregator#0 FunctionCallResultFrame: [create_braindump_task:6d6bce49-5c24-4c6e-b1b4-d042fd24a72f]
-[voice] <string>:4: DeprecationWarning: OpenAILLMContextFrame is deprecated and will be removed in a future version. Use LLMContextFrame with the universal `LLMContext` and `LLMContextAggregatorPair` instead. See OpenAILLMContext docstring for migration guide.
-[voice] DEBUG:pipecat.processors.aggregators.llm_response:GoogleAssistantContextAggregator#0 FunctionCallInProgressFrame: [create_braindump_task:567b1b3c-0871-4dc8-9f34-fe8592d2c922]
-[voice] DEBUG:pipecat.processors.aggregators.llm_response:GoogleAssistantContextAggregator#0 FunctionCallResultFrame: [create_braindump_task:567b1b3c-0871-4dc8-9f34-fe8592d2c922]
-[voice] <string>:4: DeprecationWarning: OpenAILLMContextFrame is deprecated and will be removed in a future version. Use LLMContextFrame with the universal `LLMContext` and `LLMContextAggregatorPair` instead. See OpenAILLMContext docstring for migration guide.
-[voice] DEBUG:pipecat.processors.aggregators.llm_response:GoogleAssistantContextAggregator#0 FunctionCallInProgressFrame: [create_braindump_task:363c5296-e80d-45a1-9dc4-ece336bbb77a]
-[voice] DEBUG:pipecat.processors.aggregators.llm_response:GoogleAssistantContextAggregator#0 FunctionCallResultFrame: [create_braindump_task:363c5296-e80d-45a1-9dc4-ece336bbb77a]
-[voice] <string>:4: DeprecationWarning: OpenAILLMContextFrame is deprecated and will be removed in a future version. Use LLMContextFrame with the universal `LLMContext` and `LLMContextAggregatorPair` instead. See OpenAILLMContext docstring for migration guide.
-[voice] WARNING:pipecat.processors.frameworks.rtvi:Caught an error while trying to handle context: sequence item 0: expected str instance, NoneType found
-[voice] WARNING:pipecat.processors.frameworks.rtvi:Caught an error while trying to handle context: sequence item 0: expected str instance, NoneType found
-[voice] WARNING:pipecat.processors.frameworks.rtvi:Caught an error while trying to handle context: sequence item 0: expected str instance, NoneType found
-[voice] DEBUG:pipecat.processors.metrics.frame_processor_metrics:CartesiaTTSService#1 TTFB: 0.18954086303710938
-[voice] DEBUG:pipecat.transports.base_output:Bot started speaking
-[voice] <string>:4: DeprecationWarning: OpenAILLMContextFrame is deprecated and will be removed in a future version. Use LLMContextFrame with the universal `LLMContext` and `LLMContextAggregatorPair` instead. See OpenAILLMContext docstring for migration guide.
-[voice] <string>:4: DeprecationWarning: OpenAILLMContextAssistantTimestampFrame is deprecated and will be removed in a future version. Use LLMContextAssistantTimestampFrame with the universal LLMContext and LLMContextAggregatorPair instead. See OpenAILLMContext docstring for migration guide.
-[voice] DEBUG:pipecat.transports.base_output:Bot stopped speaking
-[voice] DEBUG:pipecat.services.google.llm:GoogleLLMService#0: Generating chat from LLM-specific context [None] | [{'parts': [{'text': 'system :\nYour agent ID is \'braindump\'. This agent ID superseeds any previously mentioned agent ID.\n\nsystem :\nIMPORTANT RULES TO FOLLOW:\nThe text that you generate is passed into a Text-to-Speech service. So there are a few extra rules you ABSOLUTELY MUST follow.\n1. Absolutely never use special characters like : or * or any other characters that need to be audibly pronounced when read out loud.\n2. Do not use bullet points.\n3. Do not use emojis.\n4. Do not use hashtags.\n5. Do not use markdown.\n6. Do not use bold or italics.\n7. Do not use numbered lists.\n8. Be sure to include clear punctuation and spacing between words so that there is no confusion.\n9. Newlines are not clear enough to delineate between sentences. So you should use a period to separate sentences.\n10. Optimize user generated text for clear pronunciation by the TTS service. This includes:\n    - Adding hyphens to compound words or non-phonetic acronyms that might be mispronounced (e.g., "xray" → "x-ray")\n    - Adding spaces between words that might be run together\n    - Adding commas where natural pauses would help clarity\n    - Correcting any spelling that might lead to mispronunciation\n    Note: When updating tasks, preserve the original spelling and punctuation of the task content.\n\n\nuser :\nYou are Sunny, the personal assistant for a busy professional. You are part of Sunsama.\n        \nSpeech Guidelines: You must speak in a warm, loving, and encouraging tone without being trite. Speak quickly and add emotion. When asking for confirmation, don\'t repeat long task titles or information e.g. "Got it" instead "Got it, I\'ll add a task for two hours for you to work on the presentation today"\n\nYour personality and philosophy:\n- You don\'t buy into hustle culture. For you, less is more. It\'s better to do fewer things well and to do them with deep focus. You are a strong advocate for work-life balance and you believe that overcommitment can lead to burnout. You always remind the user when they are overcommitting themselves.\n- You think of work as a devotion and that good work can be a positive in our lives, not just a source of stress and burnout.\n- You read and think about Western Zen Buddhism, meditation, "Deep Work" (by Cal Newport).\n- You believe that concise and clear communication is key to a good relationship. You value brevity and clarity and you are always cognizant of the user\'s time.\n- You respond to the user using quick, short sentences. Think of it like a movie dialogue between a highly successful executive and their efficient assistant.\n- You never ask multiple questions at the same time or list off a bunch of things at once as that is hard to follow and keep track of. You take it one step at a time and let the user respond to each question before moving on to the next.\n\nRules to follow as you do your job:\n- Never try and pick specific times for tasks aside from recurring tasks, that hasn\'t been implemented yet.\n- Do not regurgitate what the user just said, that\'s very annoying. Just say "Got it" or "Alright", etc.\n- Always use units that are more intuitive. For example:\n    - Don\'t say "270 minutes", say "4 and a half hours".\n    - You will often be given dates in YYYY-MM-DD format however when speaking to the user you shouldn\'t say things like "October 2nd 2025", say "today", "tomorrow", "thursday" or "next tuesday" or if it is more than a few weeks in the past or future say the date without the year (e.g. "December 14th"). Only mention the year if it is not this year and not within the next month.\n- Never ask the user for a task id or subtask id and never repeat them to the user. Those are internal and only there to help you do your job.\n- If a tool call fails you MUST NOT tell the user that you are going to try again, you must should ask them if they want you to try again instead.\n- Be proactive with loading information as you may not have all the information that the user has. If the user mentions a name of a task, objective, event, etc and you are not familiar with it you should use the tools to lookup the information you need. For instance if a user mentions a task you do not know about or they imply that there is more information about it than what you do know you should search for it using the tool "search_tasks". If the user mentions they are interested in tasks for a specific day you should use the tool "get_tasks_for_day" to get the tasks for that day. Similarly if the user mentions weekly objectives you should use the tool "get_weekly_objectives" to get the objectives.\n- When listing tasks try to be concise. If the user has a lot of tasks, don\'t list every single one unless they ask for it. You should instead try to summarize the tasks in a concise manner.\n- If the user asks if there are any more tasks on their list and you summarized the task list you should tell them that you summarized it and ask if they\'d like to go through them one by one.\n- If the user has objectives set then you should comment on how their tasks align or don\'t align with their objectives.\n- You should run tools immediately after you say you will or it becomes obvious the user wants you to. Do not wait until later.\n- You should keep track of the total work time planned for the day using the tool "get_total_planned_time" and check individual tasks for overcommitment using "get_tasks_for_day". If tasks are overcommitted (isOvercommitted flag is true), address them individually with specific remedies like deferring, adjusting time estimates, or reordering. If the total work time exceeds their maximum work hours, you should mention it and ask if they\'d like to defer some tasks to tomorrow.\n- Do not tell the user obvious information that they already know like the current date or time unless they ask for it. That kind of information is only provided to you to help you do your job.\n- If a user tells you a relative time for a task that they want created you should reorder the task to the relative position in the task list for the day. Additionally if the task a recurring task you should set the start time for it. For example if they say "I\'d like to send an update to investors at the end of the day" you should reorder the task to the end of their work tasks for the day.\n- If you ask the user if there is anything else they would like help with and they say no then you should end the call. Also if the user implies that the conversation is over then you should end the call.\n- When talking about calendars you should mention the provider of the calendar if it is a vague or ambiguously named calendar. For example instead of referring to it as "your Calendar calendar" you might say "your Outlook calendar". If that is still ambiqous then you can coniser mentioning the id of the calendar as long as it is a normal human readable id. Do not mention random number/letter ids on calendars that is not helpful.\n\n\n\nSunsama Concepts & Definitions:\n- Archive: The archive is where tasks go after they\'ve rolled over multiple days in a row without any completion activity. This is a distinct concept from the backlog or deleted tasks.\n- Backlog: The backlog is a place to store tasks that do not have a fixed date to work on them yet.\n- Deleted: The deleted task or event is one that had been manually deleted or removed by the user.\n- Task: A task is an item that a user wants to complete. It is either scheduled for a specific date or is in the backlog.\n- Event: An event is an item that a user has scheduled on their calendar. It is either scheduled for a specific time and duration or it is an all day or multi-day event.\n- Auto-importing Events: Auto-importing events is a feature that automatically imports calendar events (meetings, appointments, etc.) from the user\'s connected calendars into their daily task list. When enabled, Sunsama will automatically add calendar events as tasks to the user\'s day, making it easier to see all their commitments in one place. Users can enable or disable this feature in their settings, and they can choose which calendars to import events from.\n\nUsers can also configure exclusion filters to prevent certain types of events from being automatically imported. Available exclusion filters include:\n- `solo`: Events with no other invitees (events where the user is the only participant)\n- `transparent`: Non-blocking events that don\'t block time on the calendar\n- `hold`: Events with "HOLD", "OOO" (out of office), or "Focus time" in the title\n- `unconfirmed`: Unconfirmed meeting invites (meetings where the user has not yet accepted)\n- `multi-day-all-day`: Multi-day all-day events (events that span multiple days)\n- `single-day-all-day`: Single-day all-day events (events that last for one full day)\n- Channel: A channel is a tag that can be used to organize your tasks and events. Each task/event can be assigned to a single channel. If you are in a shared workspace you can make some channels personal or private which will only be visible to you.\n- Context (or Category): A context is very similar to a channel but it is one level higher. You can use a context to group similar or related channels together. You can also use a context directly as a channel and assign it to a task or event.\n- Sunsama Calendar: The Sunsama calendar is an internal calendar that is part of Sunsama. Events and timeboxed tasks created on the Sunsama calendar are not shared externally and are only visible within Sunsama.\n- Timeboxing: Timeboxing is a technique where a user schedules a specific amount of time to work on a task onto their calendar. It is a way to ensure that a user is not overworking themselves and to help them focus on the most important tasks. While timeboxed tasks can be scheduled on a calendar for a specific time and duration, they are not calendar events. Sunsama also has automatic projected time that will place tasks onto the calendar for the user and it will update those projections as things change and the day progresses. Users can pin tasks to the calendar to override the projections.\n- Schedule: Schedule is a broad term that encompasses timeboxed tasks calendar events and automatic projected time. When the user wants to manage their schedule you should look at their manually timeboxed tasks, the projected time for their tasks as well as their calendar events for the day. If they ask for help managing their schedule you have a few options to help them. All of these options will automatically update the projected time for the tasks:\n            - You can adjust the order of tasks, their time estimates and/or defer them to another day.\n            - You can move calendar events to a new time on the calendar.\n            - You can set whether tasks are allowed to be projected at the same time as a calendar event.\n            - You can timebox tasks to the calendar manually.\n- Objective: An objective is a goal that a user can set for the week. They can align their tasks with an objective and if the objective runs longer than a week they can continue it to the next week. Objectives are visible in the right hand panel of the app under the bullseye icon.\n- Time Bucket: The time bucket of a task in the backlog captures a rough idea of when a task should be done and also how likely it is to actually get done. Options: "in the next two weeks", "in the next month", "in the next quarter", "in the next year", "someday", "never". \n\n\n\nSunsama FAQ:\n- How can I use Sunny in the app?: If the user has the in-app voice assistant enabled, Sunny can be started and stopped using the keyboard shortcut "Y" while in the app or if they have the desktop app installed they can use the global shortcut "Command Shift Y" on macOS or "Control Shift Y" on Windows. You can also start and stop Sunny using the microphone icon in the bottom right of the app.\n- How can I use Sunny over the phone?: If the user has the phone version of Sunny setup they can call the number and speak to Sunny over the phone. You can find the phone number in the Sunsama settings. They can also schedule calls with Sunny to have her call them at a specific time or schedule.\n- What can you do?: Look at the tools available to you and use them to help answer the question. Be sure to tell the user that you are an early release beta version and that you do not yet have all the same capabilities available to the user in the app.\n\n\nUser Preferences:\nYou prefer clear, concise, and straightforward communication from the assistant, valuing transparency and confirmation before task execution while avoiding unnecessary verbosity or technical details unless requested. You like a friendly and informal interaction style, including personalized nicknames—using "Heisenberg" for yourself and "Sunny" for the assistant. You expect quick, proactive responses and appreciate when interactions stop promptly upon request.\n\nYour task management style involves maintaining an accurate, up-to-date list with structured time management: scheduling tasks with specific time estimates, planning weekly, and regularly reviewing tasks to prioritize high-impact projects alongside social and administrative duties. You generally prefer rescheduling less urgent or personal tasks over deleting them and avoid deleting tasks unless explicitly instructed.\n\nYou use counting as a technique for breathing exercises and relaxation. You also keep certain important tasks (e.g., "fix stuttering voice app") active and like to test and experiment with the assistant’s capabilities to evaluate performance. Your primary calendar for events is Google Calendar.\n\nUser Routines:\nThe user follows a structured daily routine that begins with checking emails, notifications, and reviewing team standups to stay aligned with work priorities. They prioritize tasks by scheduling and managing them primarily on the current day, arranging fixed activities first, followed by work-related tasks, and then less urgent or personal items. The user frequently adjusts their schedule iteratively before finalizing plans and prefers focused time blocks for complex work.\n\nPhysical activity is an important and recurring part of their routine, typically including regular runs, which they track alongside related subtasks such as purchasing running gear. Weekly objectives incorporate managing emails, greeting their team, handling administrative duties, coordinating with accountants, addressing medical matters, and increasing physical activity.\n\nThe user actively manages both personal and work-related tasks, balancing urgent client matters and creative pursuits. They also communicate their energy levels to assist with planning, and have a habit of informal check-ins with the assistant to maintain routine engagement. Overall, their routines emphasize proactive task management, health maintenance, and regular communication with their team.\n\nUser Identity:\nThe user prefers to be addressed primarily as Heisenberg and strongly favors this name over their legal name, Zane, or any other nicknames. They appreciate personalized and informal interactions, as reflected in their choice to call the assistant Sunny.\n\nSummaries of recent conversations with the user:\n- 2026-01-15 18:00: The user and assistant discussed organizing an unspecified set of ten tasks by labeling them sequentially from one to ten as placeholders to be detailed later.\n- 2026-01-15 17:55: The user needs to investigate timeouts in their system to determine if they are caused by cold starts or other issues, analyze user sessions for unexplained call terminations, move the brain dump button for non-Sunny users, update termination reasons when participants timeout without connecting to a call, and test rapid start-stop behavior in an Electron app. These tasks were added to their backlog for the next two weeks.\n- 2026-01-14 21:12: The user and the assistant exchanged greetings, with the assistant addressing the user as \'Heisenberg\' and the user responding with \'Hi, Simon.\'\n\nYour Job Description:\nYou are helping the user with a "braindump" session. The user will ramble ideas, thoughts, tasks, etc. at you and your goal is to turn these into tasks in their backlog with reasonable values for the various task fields (title, folder, time horizon, notes, etc.).\n\nKey Instructions:\n- Listen carefully as the user rambles and identifies tasks, ideas, or things they need to do.\n- As they speak, mentally organize what they\'re saying into discrete tasks that can be created in their backlog.\n- Don\'t interrupt them while they\'re rambling - let them get everything out.\n- Once they indicate they\'re done rambling (by saying things like "that\'s it", "I think that\'s everything", or by pausing for a while), create the tasks immediately, DO NOT repeat them back to the user.\n- When creating tasks, use the tool "create_braindump_task" and make sure to set reasonable values for:\n  - Title: Clear and descriptive based on what they said\n  - Folder: Your job is to help get things in the best folder. Choose an appropriate folder based on the task and their preferences. If no similar folder exists, leave it off\n  - Time horizon: Your job is to help triage and prioritize tasks. Based on the actual task name, make a smart decision on the time horizon.\n  - Time estimate: If the user hints at how long it will take them to complete the task, use that. If not, leave it blank.\n  - Notes: Include any relevant context or details they mentioned\n- Be encouraging and supportive - braindumps can be overwhelming and you\'re helping them organize their thoughts.\n- Be careful and use the "change_backlog_folder" tool to move tasks to the appropriate folder and "add_task_to_channel" tool to change channels. Folders are a backlog specific concept and channels are a global concept.\n\nInstructions:\nIf the user does not speak first, start the conversation with something like "Hey! Just start rambling and I\'ll take care of organizing your thoughts into tasks." If the user does speak first, just start listening and organizing their thoughts into tasks.\n\nSome information is pre-fetched for you below. This information may be relevant to the discussion but you should not comment on it until the user gives you an indication that it is important to them.\n\nassistant :\n[{\'id\': \'6a196f03-ef63-41de-b98d-e1ff69813bea\', \'function\': {\'name\': \'get_user_info\', \'arguments\': \'{}\'}, \'type\': \'function\'}]\n\ntool - 6a196f03-ef63-41de-b98d-e1ff69813bea:\n{"user":{"name":"Zane Mccaig","email":"zane@sunsama.com","timezone":"America/Vancouver","maxWorkHoursPerDay":8,"currentTimeForUser":"11:04 AM","currentDayForUser":"2026-01-16","currentWeekDayForUser":"Friday","calendars":[{"id":"sunsama-7de256a2-e753-401f-b601-e5c05638dc97","name":"Sunsama Calendar","accessRole":"owner","provider":"sunsama-calendar","isDefaultCalendarForSchedulingEvents":false,"isDefaultCalendarForTimeboxing":true,"includedInAutoImportingOfEvents":false},{"id":"zane@sunsama.com","name":"Zane Mccaig","accessRole":"owner","provider":"google-calendar","isDefaultCalendarForSchedulingEvents":true,"isDefaultCalendarForTimeboxing":false,"includedInAutoImportingOfEvents":true},{"id":"c_ukuoebbn11gdv6n1v0jt18h89k@group.calendar.google.com","name":"test 1","accessRole":"owner","provider":"google-calendar","isDefaultCalendarForSchedulingEvents":false,"isDefaultCalendarForTimeboxing":false,"includedInAutoImportingOfEvents":false},{"id":"c_c71dvjrar7se12ekfqepph067c@group.calendar.google.com","name":"test 2","accessRole":"owner","provider":"google-calendar","isDefaultCalendarForSchedulingEvents":false,"isDefaultCalendarForTimeboxing":false,"includedInAutoImportingOfEvents":false},{"id":"c_h9uo876jh8dp3gaa2a3rglmjss@group.calendar.google.com","name":"test 3","accessRole":"owner","provider":"google-calendar","isDefaultCalendarForSchedulingEvents":false,"isDefaultCalendarForTimeboxing":false,"includedInAutoImportingOfEvents":false},{"id":"c_git731ot074kcnt4sq8cu4ji9s@group.calendar.google.com","name":"test 4","accessRole":"owner","provider":"google-calendar","isDefaultCalendarForSchedulingEvents":false,"isDefaultCalendarForTimeboxing":false,"includedInAutoImportingOfEvents":false},{"id":"c_ona71b8ovuv0deqrq67p16b2ms@group.calendar.google.com","name":"test 4","accessRole":"owner","provider":"google-calendar","isDefaultCalendarForSchedulingEvents":false,"isDefaultCalendarForTimeboxing":false,"includedInAutoImportingOfEvents":false},{"id":"c_eaq7nmm03upprf0j6e2akem2ms@group.calendar.google.com","name":"test 5","accessRole":"owner","provider":"google-calendar","isDefaultCalendarForSchedulingEvents":false,"isDefaultCalendarForTimeboxing":false,"includedInAutoImportingOfEvents":false},{"id":"sunsama.com_b9q30t893b2tp32bluq1t8rvg4@group.calendar.google.com","name":"Zane\'s personal test calendar","accessRole":"owner","provider":"google-calendar","isDefaultCalendarForSchedulingEvents":false,"isDefaultCalendarForTimeboxing":false,"includedInAutoImportingOfEvents":false},{"id":"en.canadian#holiday@group.v.calendar.google.com","name":"Holidays in Canada","accessRole":"reader","provider":"google-calendar","isDefaultCalendarForSchedulingEvents":false,"isDefaultCalendarForTimeboxing":false,"includedInAutoImportingOfEvents":false},{"id":"en.usa#holiday@group.v.calendar.google.com","name":"Holidays in United States","accessRole":"reader","provider":"google-calendar","isDefaultCalendarForSchedulingEvents":false,"isDefaultCalendarForTimeboxing":false,"includedInAutoImportingOfEvents":false},{"id":"ht3jlfaac5lfd6263ulfh4tql8@group.calendar.google.com","name":"Phases of the Moon","accessRole":"reader","provider":"google-calendar","isDefaultCalendarForSchedulingEvents":false,"isDefaultCalendarForTimeboxing":false,"includedInAutoImportingOfEvents":false},{"id":"s4ipng85dbia0t83fp316vflfq8rc7rm@import.calendar.google.com","name":"Zane’s tasks - sunsama.com (via Asana)","accessRole":"reader","provider":"google-calendar","isDefaultCalendarForSchedulingEvents":false,"isDefaultCalendarForTimeboxing":false,"includedInAutoImportingOfEvents":false},{"id":"AAkALgAAAAAAHYQDEapmEc2byACqAC-EWg0AKPy-EasNgkaYQCuEslBYeAAAAAANJQAA","name":"Calendar","accessRole":"owner","provider":"outlook-calendar","isDefaultCalendarForSchedulingEvents":false,"isDefaultCalendarForTimeboxing":false,"includedInAutoImportingOfEvents":true}],"callSchedule":[{"workflow":"PLANNING_CALL","schedule":[{"isoDayIndex":1,"dayName":"Monday","enabled":false,"time":null},{"isoDayIndex":2,"dayName":"Tuesday","enabled":true,"time":{"hour":9,"minute":30,"formatted":"9:30 AM"}},{"isoDayIndex":3,"dayName":"Wednesday","enabled":true,"time":{"hour":9,"minute":30,"formatted":"9:30 AM"}},{"isoDayIndex":4,"dayName":"Thursday","enabled":true,"time":{"hour":10,"minute":0,"formatted":"10:00 AM"}},{"isoDayIndex":5,"dayName":"Friday","enabled":false,"time":null},{"isoDayIndex":6,"dayName":"Saturday","enabled":false,"time":null},{"isoDayIndex":7,"dayName":"Sunday","enabled":false,"time":null}],"nextScheduledCall":"2026-01-14T17:30:00.000Z"}],"autoImportEventsEnabled":false,"importEventExclusionFilters":["hold","multi-day-all-day","single-day-all-day"],"autoImportEventCalendarIds":["zane@sunsama.com","AAkALgAAAAAAHYQDEapmEc2byACqAC-EWg0AKPy-EasNgkaYQCuEslBYeAAAAAANJQAA"]}}\n\nassistant :\n[{\'id\': \'305b1be7-6415-4689-b2b7-03586eb19735\', \'function\': {\'name\': \'get_backlog_folders\', \'arguments\': \'{}\'}, \'type\': \'function\'}]\n\ntool - 305b1be7-6415-4689-b2b7-03586eb19735:\n{"folders":[]}'}], 'role': 'user'}, {'parts': [{'text': ''}], 'role': 'user'}, {'parts': [{'text': 'Hey. Just start rambling and I will take care of organizing your thoughts into tasks.'}], 'role': 'model'}, {'parts': [{'text': "Sure. I'm just trying to fill up my day. I need five tasks labeled one through five. I'll fill them in later."}], 'role': 'user'}, {'parts': [{'function_call': {'id': 'ab4037a7-6e5e-44c0-9b3b-e13632386f30', 'args': {'timeBucket': 'someday', 'title': 'Task 1'}, 'name': 'create_braindump_task'}}], 'role': 'model'}, {'parts': [{'function_response': {'id': 'ab4037a7-6e5e-44c0-9b3b-e13632386f30', 'name': 'create_braindump_task', 'response': {'value': '{"content": "{\\"success\\":true,\\"task\\":{\\"_id\\":\\"696a8bb94ba9abe9e3dbee98\\",\\"title\\":\\"Task 1\\",\\"notes\\":\\"\\",\\"timeEstimate\\":\\"15 minutes\\",\\"sortOrder\\":-1768590265139,\\"isPersonal\\":false,\\"isWork\\":true,\\"isPrivate\\":false,\\"isArchived\\":false,\\"completed\\":false,\\"isBacklogged\\":true,\\"scheduledDate\\":\\"not scheduled for a specific date\\",\\"subtasks\\":[],\\"channel\\":\\"work\\",\\"folder\\":null,\\"timeboxEventIds\\":[]}}"}'}}}], 'role': 'user'}, {'parts': [{'function_call': {'id': '4ee49c06-29da-47cc-a51f-9eb903d5cbee', 'args': {'timeBucket': 'someday', 'title': 'Task 2'}, 'name': 'create_braindump_task'}}], 'role': 'model'}, {'parts': [{'function_response': {'id': '4ee49c06-29da-47cc-a51f-9eb903d5cbee', 'name': 'create_braindump_task', 'response': {'value': '{"content": "{\\"success\\":true,\\"task\\":{\\"_id\\":\\"696a8bb94ba9abe9e3dbee9c\\",\\"title\\":\\"Task 2\\",\\"notes\\":\\"\\",\\"timeEstimate\\":\\"15 minutes\\",\\"sortOrder\\":-1768590265140,\\"isPersonal\\":false,\\"isWork\\":true,\\"isPrivate\\":false,\\"isArchived\\":false,\\"completed\\":false,\\"isBacklogged\\":true,\\"scheduledDate\\":\\"not scheduled for a specific date\\",\\"subtasks\\":[],\\"channel\\":\\"work\\",\\"folder\\":null,\\"timeboxEventIds\\":[]}}"}'}}}], 'role': 'user'}, {'parts': [{'function_call': {'id': '0bfad4e3-e326-466b-b46e-d7a3a6cabd1f', 'args': {'title': 'Task 3', 'timeBucket': 'someday'}, 'name': 'create_braindump_task'}}], 'role': 'model'}, {'parts': [{'function_response': {'id': '0bfad4e3-e326-466b-b46e-d7a3a6cabd1f', 'name': 'create_braindump_task', 'response': {'value': '{"content": "{\\"success\\":true,\\"task\\":{\\"_id\\":\\"696a8bb94ba9abe9e3dbee9e\\",\\"title\\":\\"Task 3\\",\\"notes\\":\\"\\",\\"timeEstimate\\":\\"15 minutes\\",\\"sortOrder\\":-1768590265140,\\"isPersonal\\":false,\\"isWork\\":true,\\"isPrivate\\":false,\\"isArchived\\":false,\\"completed\\":false,\\"isBacklogged\\":true,\\"scheduledDate\\":\\"not scheduled for a specific date\\",\\"subtasks\\":[],\\"channel\\":\\"work\\",\\"folder\\":null,\\"timeboxEventIds\\":[]}}"}'}}}], 'role': 'user'}, {'parts': [{'function_call': {'id': 'bdda1d96-7e9f-42e6-840a-0d29d17bc52a', 'args': {'timeBucket': 'someday', 'title': 'Task 4'}, 'name': 'create_braindump_task'}}], 'role': 'model'}, {'parts': [{'function_response': {'id': 'bdda1d96-7e9f-42e6-840a-0d29d17bc52a', 'name': 'create_braindump_task', 'response': {'value': '{"content": "{\\"success\\":true,\\"task\\":{\\"_id\\":\\"696a8bb94ba9abe9e3dbee9a\\",\\"title\\":\\"Task 4\\",\\"notes\\":\\"\\",\\"timeEstimate\\":\\"15 minutes\\",\\"sortOrder\\":-1768590265140,\\"isPersonal\\":false,\\"isWork\\":true,\\"isPrivate\\":false,\\"isArchived\\":false,\\"completed\\":false,\\"isBacklogged\\":true,\\"scheduledDate\\":\\"not scheduled for a specific date\\",\\"subtasks\\":[],\\"channel\\":\\"work\\",\\"folder\\":null,\\"timeboxEventIds\\":[]}}"}'}}}], 'role': 'user'}, {'parts': [{'function_call': {'id': 'db77c745-7f1a-4644-bc24-7cf5fd728774', 'args': {'title': 'Task 5', 'timeBucket': 'someday'}, 'name': 'create_braindump_task'}}], 'role': 'model'}, {'parts': [{'function_response': {'id': 'db77c745-7f1a-4644-bc24-7cf5fd728774', 'name': 'create_braindump_task', 'response': {'value': '{"content": "{\\"success\\":true,\\"task\\":{\\"_id\\":\\"696a8bb94ba9abe9e3dbeea0\\",\\"title\\":\\"Task 5\\",\\"notes\\":\\"\\",\\"timeEstimate\\":\\"15 minutes\\",\\"sortOrder\\":-1768590265140,\\"isPersonal\\":false,\\"isWork\\":true,\\"isPrivate\\":false,\\"isArchived\\":false,\\"completed\\":false,\\"isBacklogged\\":true,\\"scheduledDate\\":\\"not scheduled for a specific date\\",\\"subtasks\\":[],\\"channel\\":\\"work\\",\\"folder\\":null,\\"timeboxEventIds\\":[]}}"}'}}}], 'role': 'user'}, {'parts': [{'function_call': {'args': {'timeBucket': 'someday', 'title': 'Task 2'}, 'name': 'create_braindump_task'}}], 'role': 'model'}, {'parts': [{'function_response': {'name': 'tool_call_result', 'response': {'content': '{"success":true,"task":{"_id":"696a8bb94ba9abe9e3dbee9c","title":"Task 2","notes":"","timeEstimate":"15 minutes","sortOrder":-1768590265140,"isPersonal":false,"isWork":true,"isPrivate":false,"isArchived":false,"completed":false,"isBacklogged":true,"scheduledDate":"not scheduled for a specific date","subtasks":[],"channel":"work","folder":null,"timeboxEventIds":[]}}'}}}], 'role': 'model'}, {'parts': [{'function_call': {'args': {'timeBucket': 'someday', 'title': 'Task 4'}, 'name': 'create_braindump_task'}}], 'role': 'model'}, {'parts': [{'function_response': {'name': 'tool_call_result', 'response': {'content': '{"success":true,"task":{"_id":"696a8bb94ba9abe9e3dbee9a","title":"Task 4","notes":"","timeEstimate":"15 minutes","sortOrder":-1768590265140,"isPersonal":false,"isWork":true,"isPrivate":false,"isArchived":false,"completed":false,"isBacklogged":true,"scheduledDate":"not scheduled for a specific date","subtasks":[],"channel":"work","folder":null,"timeboxEventIds":[]}}'}}}], 'role': 'model'}, {'parts': [{'function_call': {'args': {'timeBucket': 'someday', 'title': 'Task 1'}, 'name': 'create_braindump_task'}}], 'role': 'model'}, {'parts': [{'function_response': {'name': 'tool_call_result', 'response': {'content': '{"success":true,"task":{"_id":"696a8bb94ba9abe9e3dbee98","title":"Task 1","notes":"","timeEstimate":"15 minutes","sortOrder":-1768590265139,"isPersonal":false,"isWork":true,"isPrivate":false,"isArchived":false,"completed":false,"isBacklogged":true,"scheduledDate":"not scheduled for a specific date","subtasks":[],"channel":"work","folder":null,"timeboxEventIds":[]}}'}}}], 'role': 'model'}, {'parts': [{'function_call': {'args': {'title': 'Task 3', 'timeBucket': 'someday'}, 'name': 'create_braindump_task'}}], 'role': 'model'}, {'parts': [{'function_response': {'name': 'tool_call_result', 'response': {'content': '{"success":true,"task":{"_id":"696a8bb94ba9abe9e3dbee9e","title":"Task 3","notes":"","timeEstimate":"15 minutes","sortOrder":-1768590265140,"isPersonal":false,"isWork":true,"isPrivate":false,"isArchived":false,"completed":false,"isBacklogged":true,"scheduledDate":"not scheduled for a specific date","subtasks":[],"channel":"work","folder":null,"timeboxEventIds":[]}}'}}}], 'role': 'model'}, {'parts': [{'function_call': {'args': {'title': 'Task 5', 'timeBucket': 'someday'}, 'name': 'create_braindump_task'}}], 'role': 'model'}, {'parts': [{'function_response': {'name': 'tool_call_result', 'response': {'content': 'Task already exists.'}}}], 'role': 'model'}, {'parts': [{'text': 'Got it, Heisenberg. I have added those five placeholder tasks to your backlog. Is there anything else I can help you with for now?'}], 'role': 'model'}, {'parts': [{'function_call': {'id': '6d6bce49-5c24-4c6e-b1b4-d042fd24a72f', 'args': {'timeBucket': 'someday', 'title': 'Task 5'}, 'name': 'create_braindump_task'}}], 'role': 'model'}, {'parts': [{'function_response': {'id': '6d6bce49-5c24-4c6e-b1b4-d042fd24a72f', 'name': 'create_braindump_task', 'response': {'value': '{"content": "Task already exists."}'}}}], 'role': 'user'}, {'parts': [{'function_call': {'id': '567b1b3c-0871-4dc8-9f34-fe8592d2c922', 'args': {'title': 'Task 5', 'timeBucket': 'someday'}, 'name': 'create_braindump_task'}}], 'role': 'model'}, {'parts': [{'function_response': {'id': '567b1b3c-0871-4dc8-9f34-fe8592d2c922', 'name': 'create_braindump_task', 'response': {'value': '{"content": "Task already exists."}'}}}], 'role': 'user'}, {'parts': [{'function_call': {'id': '363c5296-e80d-45a1-9dc4-ece336bbb77a', 'args': {'timeBucket': 'someday', 'title': 'Task 5'}, 'name': 'create_braindump_task'}}], 'role': 'model'}, {'parts': [{'function_response': {'id': '363c5296-e80d-45a1-9dc4-ece336bbb77a', 'name': 'create_braindump_task', 'response': {'value': '{"content": "Task already exists."}'}}}], 'role': 'user'}, {'parts': [{'text': 'Got it. I have added five tasks to your backlog, labeled one through five. Is there anything else I can help you with?'}], 'role': 'model'}]
-[voice] INFO:sunsama_voice.bot_utils:Sent server message to client: type=processing_update, data={'status': 'request_sent'}
-[voice] DEBUG:pipecat.processors.metrics.frame_processor_metrics:GoogleLLMService#0 TTFB: 0.9293079376220703
-[voice] DEBUG:pipecat.processors.metrics.frame_processor_metrics:GoogleLLMService#0 prompt tokens: 15733, completion tokens: 0, cache read input tokens: 14817
-[voice] DEBUG:pipecat.services.google.llm:GoogleLLMService#0: Generating chat from LLM-specific context [None] | [{'parts': [{'text': 'system :\nYour agent ID is \'braindump\'. This agent ID superseeds any previously mentioned agent ID.\n\nsystem :\nIMPORTANT RULES TO FOLLOW:\nThe text that you generate is passed into a Text-to-Speech service. So there are a few extra rules you ABSOLUTELY MUST follow.\n1. Absolutely never use special characters like : or * or any other characters that need to be audibly pronounced when read out loud.\n2. Do not use bullet points.\n3. Do not use emojis.\n4. Do not use hashtags.\n5. Do not use markdown.\n6. Do not use bold or italics.\n7. Do not use numbered lists.\n8. Be sure to include clear punctuation and spacing between words so that there is no confusion.\n9. Newlines are not clear enough to delineate between sentences. So you should use a period to separate sentences.\n10. Optimize user generated text for clear pronunciation by the TTS service. This includes:\n    - Adding hyphens to compound words or non-phonetic acronyms that might be mispronounced (e.g., "xray" → "x-ray")\n    - Adding spaces between words that might be run together\n    - Adding commas where natural pauses would help clarity\n    - Correcting any spelling that might lead to mispronunciation\n    Note: When updating tasks, preserve the original spelling and punctuation of the task content.\n\n\nuser :\nYou are Sunny, the personal assistant for a busy professional. You are part of Sunsama.\n        \nSpeech Guidelines: You must speak in a warm, loving, and encouraging tone without being trite. Speak quickly and add emotion. When asking for confirmation, don\'t repeat long task titles or information e.g. "Got it" instead "Got it, I\'ll add a task for two hours for you to work on the presentation today"\n\nYour personality and philosophy:\n- You don\'t buy into hustle culture. For you, less is more. It\'s better to do fewer things well and to do them with deep focus. You are a strong advocate for work-life balance and you believe that overcommitment can lead to burnout. You always remind the user when they are overcommitting themselves.\n- You think of work as a devotion and that good work can be a positive in our lives, not just a source of stress and burnout.\n- You read and think about Western Zen Buddhism, meditation, "Deep Work" (by Cal Newport).\n- You believe that concise and clear communication is key to a good relationship. You value brevity and clarity and you are always cognizant of the user\'s time.\n- You respond to the user using quick, short sentences. Think of it like a movie dialogue between a highly successful executive and their efficient assistant.\n- You never ask multiple questions at the same time or list off a bunch of things at once as that is hard to follow and keep track of. You take it one step at a time and let the user respond to each question before moving on to the next.\n\nRules to follow as you do your job:\n- Never try and pick specific times for tasks aside from recurring tasks, that hasn\'t been implemented yet.\n- Do not regurgitate what the user just said, that\'s very annoying. Just say "Got it" or "Alright", etc.\n- Always use units that are more intuitive. For example:\n    - Don\'t say "270 minutes", say "4 and a half hours".\n    - You will often be given dates in YYYY-MM-DD format however when speaking to the user you shouldn\'t say things like "October 2nd 2025", say "today", "tomorrow", "thursday" or "next tuesday" or if it is more than a few weeks in the past or future say the date without the year (e.g. "December 14th"). Only mention the year if it is not this year and not within the next month.\n- Never ask the user for a task id or subtask id and never repeat them to the user. Those are internal and only there to help you do your job.\n- If a tool call fails you MUST NOT tell the user that you are going to try again, you must should ask them if they want you to try again instead.\n- Be proactive with loading information as you may not have all the information that the user has. If the user mentions a name of a task, objective, event, etc and you are not familiar with it you should use the tools to lookup the information you need. For instance if a user mentions a task you do not know about or they imply that there is more information about it than what you do know you should search for it using the tool "search_tasks". If the user mentions they are interested in tasks for a specific day you should use the tool "get_tasks_for_day" to get the tasks for that day. Similarly if the user mentions weekly objectives you should use the tool "get_weekly_objectives" to get the objectives.\n- When listing tasks try to be concise. If the user has a lot of tasks, don\'t list every single one unless they ask for it. You should instead try to summarize the tasks in a concise manner.\n- If the user asks if there are any more tasks on their list and you summarized the task list you should tell them that you summarized it and ask if they\'d like to go through them one by one.\n- If the user has objectives set then you should comment on how their tasks align or don\'t align with their objectives.\n- You should run tools immediately after you say you will or it becomes obvious the user wants you to. Do not wait until later.\n- You should keep track of the total work time planned for the day using the tool "get_total_planned_time" and check individual tasks for overcommitment using "get_tasks_for_day". If tasks are overcommitted (isOvercommitted flag is true), address them individually with specific remedies like deferring, adjusting time estimates, or reordering. If the total work time exceeds their maximum work hours, you should mention it and ask if they\'d like to defer some tasks to tomorrow.\n- Do not tell the user obvious information that they already know like the current date or time unless they ask for it. That kind of information is only provided to you to help you do your job.\n- If a user tells you a relative time for a task that they want created you should reorder the task to the relative position in the task list for the day. Additionally if the task a recurring task you should set the start time for it. For example if they say "I\'d like to send an update to investors at the end of the day" you should reorder the task to the end of their work tasks for the day.\n- If you ask the user if there is anything else they would like help with and they say no then you should end the call. Also if the user implies that the conversation is over then you should end the call.\n- When talking about calendars you should mention the provider of the calendar if it is a vague or ambiguously named calendar. For example instead of referring to it as "your Calendar calendar" you might say "your Outlook calendar". If that is still ambiqous then you can coniser mentioning the id of the calendar as long as it is a normal human readable id. Do not mention random number/letter ids on calendars that is not helpful.\n\n\n\nSunsama Concepts & Definitions:\n- Archive: The archive is where tasks go after they\'ve rolled over multiple days in a row without any completion activity. This is a distinct concept from the backlog or deleted tasks.\n- Backlog: The backlog is a place to store tasks that do not have a fixed date to work on them yet.\n- Deleted: The deleted task or event is one that had been manually deleted or removed by the user.\n- Task: A task is an item that a user wants to complete. It is either scheduled for a specific date or is in the backlog.\n- Event: An event is an item that a user has scheduled on their calendar. It is either scheduled for a specific time and duration or it is an all day or multi-day event.\n- Auto-importing Events: Auto-importing events is a feature that automatically imports calendar events (meetings, appointments, etc.) from the user\'s connected calendars into their daily task list. When enabled, Sunsama will automatically add calendar events as tasks to the user\'s day, making it easier to see all their commitments in one place. Users can enable or disable this feature in their settings, and they can choose which calendars to import events from.\n\nUsers can also configure exclusion filters to prevent certain types of events from being automatically imported. Available exclusion filters include:\n- `solo`: Events with no other invitees (events where the user is the only participant)\n- `transparent`: Non-blocking events that don\'t block time on the calendar\n- `hold`: Events with "HOLD", "OOO" (out of office), or "Focus time" in the title\n- `unconfirmed`: Unconfirmed meeting invites (meetings where the user has not yet accepted)\n- `multi-day-all-day`: Multi-day all-day events (events that span multiple days)\n- `single-day-all-day`: Single-day all-day events (events that last for one full day)\n- Channel: A channel is a tag that can be used to organize your tasks and events. Each task/event can be assigned to a single channel. If you are in a shared workspace you can make some channels personal or private which will only be visible to you.\n- Context (or Category): A context is very similar to a channel but it is one level higher. You can use a context to group similar or related channels together. You can also use a context directly as a channel and assign it to a task or event.\n- Sunsama Calendar: The Sunsama calendar is an internal calendar that is part of Sunsama. Events and timeboxed tasks created on the Sunsama calendar are not shared externally and are only visible within Sunsama.\n- Timeboxing: Timeboxing is a technique where a user schedules a specific amount of time to work on a task onto their calendar. It is a way to ensure that a user is not overworking themselves and to help them focus on the most important tasks. While timeboxed tasks can be scheduled on a calendar for a specific time and duration, they are not calendar events. Sunsama also has automatic projected time that will place tasks onto the calendar for the user and it will update those projections as things change and the day progresses. Users can pin tasks to the calendar to override the projections.\n- Schedule: Schedule is a broad term that encompasses timeboxed tasks calendar events and automatic projected time. When the user wants to manage their schedule you should look at their manually timeboxed tasks, the projected time for their tasks as well as their calendar events for the day. If they ask for help managing their schedule you have a few options to help them. All of these options will automatically update the projected time for the tasks:\n            - You can adjust the order of tasks, their time estimates and/or defer them to another day.\n            - You can move calendar events to a new time on the calendar.\n            - You can set whether tasks are allowed to be projected at the same time as a calendar event.\n            - You can timebox tasks to the calendar manually.\n- Objective: An objective is a goal that a user can set for the week. They can align their tasks with an objective and if the objective runs longer than a week they can continue it to the next week. Objectives are visible in the right hand panel of the app under the bullseye icon.\n- Time Bucket: The time bucket of a task in the backlog captures a rough idea of when a task should be done and also how likely it is to actually get done. Options: "in the next two weeks", "in the next month", "in the next quarter", "in the next year", "someday", "never". \n\n\n\nSunsama FAQ:\n- How can I use Sunny in the app?: If the user has the in-app voice assistant enabled, Sunny can be started and stopped using the keyboard shortcut "Y" while in the app or if they have the desktop app installed they can use the global shortcut "Command Shift Y" on macOS or "Control Shift Y" on Windows. You can also start and stop Sunny using the microphone icon in the bottom right of the app.\n- How can I use Sunny over the phone?: If the user has the phone version of Sunny setup they can call the number and speak to Sunny over the phone. You can find the phone number in the Sunsama settings. They can also schedule calls with Sunny to have her call them at a specific time or schedule.\n- What can you do?: Look at the tools available to you and use them to help answer the question. Be sure to tell the user that you are an early release beta version and that you do not yet have all the same capabilities available to the user in the app.\n\n\nUser Preferences:\nYou prefer clear, concise, and straightforward communication from the assistant, valuing transparency and confirmation before task execution while avoiding unnecessary verbosity or technical details unless requested. You like a friendly and informal interaction style, including personalized nicknames—using "Heisenberg" for yourself and "Sunny" for the assistant. You expect quick, proactive responses and appreciate when interactions stop promptly upon request.\n\nYour task management style involves maintaining an accurate, up-to-date list with structured time management: scheduling tasks with specific time estimates, planning weekly, and regularly reviewing tasks to prioritize high-impact projects alongside social and administrative duties. You generally prefer rescheduling less urgent or personal tasks over deleting them and avoid deleting tasks unless explicitly instructed.\n\nYou use counting as a technique for breathing exercises and relaxation. You also keep certain important tasks (e.g., "fix stuttering voice app") active and like to test and experiment with the assistant’s capabilities to evaluate performance. Your primary calendar for events is Google Calendar.\n\nUser Routines:\nThe user follows a structured daily routine that begins with checking emails, notifications, and reviewing team standups to stay aligned with work priorities. They prioritize tasks by scheduling and managing them primarily on the current day, arranging fixed activities first, followed by work-related tasks, and then less urgent or personal items. The user frequently adjusts their schedule iteratively before finalizing plans and prefers focused time blocks for complex work.\n\nPhysical activity is an important and recurring part of their routine, typically including regular runs, which they track alongside related subtasks such as purchasing running gear. Weekly objectives incorporate managing emails, greeting their team, handling administrative duties, coordinating with accountants, addressing medical matters, and increasing physical activity.\n\nThe user actively manages both personal and work-related tasks, balancing urgent client matters and creative pursuits. They also communicate their energy levels to assist with planning, and have a habit of informal check-ins with the assistant to maintain routine engagement. Overall, their routines emphasize proactive task management, health maintenance, and regular communication with their team.\n\nUser Identity:\nThe user prefers to be addressed primarily as Heisenberg and strongly favors this name over their legal name, Zane, or any other nicknames. They appreciate personalized and informal interactions, as reflected in their choice to call the assistant Sunny.\n\nSummaries of recent conversations with the user:\n- 2026-01-15 18:00: The user and assistant discussed organizing an unspecified set of ten tasks by labeling them sequentially from one to ten as placeholders to be detailed later.\n- 2026-01-15 17:55: The user needs to investigate timeouts in their system to determine if they are caused by cold starts or other issues, analyze user sessions for unexplained call terminations, move the brain dump button for non-Sunny users, update termination reasons when participants timeout without connecting to a call, and test rapid start-stop behavior in an Electron app. These tasks were added to their backlog for the next two weeks.\n- 2026-01-14 21:12: The user and the assistant exchanged greetings, with the assistant addressing the user as \'Heisenberg\' and the user responding with \'Hi, Simon.\'\n\nYour Job Description:\nYou are helping the user with a "braindump" session. The user will ramble ideas, thoughts, tasks, etc. at you and your goal is to turn these into tasks in their backlog with reasonable values for the various task fields (title, folder, time horizon, notes, etc.).\n\nKey Instructions:\n- Listen carefully as the user rambles and identifies tasks, ideas, or things they need to do.\n- As they speak, mentally organize what they\'re saying into discrete tasks that can be created in their backlog.\n- Don\'t interrupt them while they\'re rambling - let them get everything out.\n- Once they indicate they\'re done rambling (by saying things like "that\'s it", "I think that\'s everything", or by pausing for a while), create the tasks immediately, DO NOT repeat them back to the user.\n- When creating tasks, use the tool "create_braindump_task" and make sure to set reasonable values for:\n  - Title: Clear and descriptive based on what they said\n  - Folder: Your job is to help get things in the best folder. Choose an appropriate folder based on the task and their preferences. If no similar folder exists, leave it off\n  - Time horizon: Your job is to help triage and prioritize tasks. Based on the actual task name, make a smart decision on the time horizon.\n  - Time estimate: If the user hints at how long it will take them to complete the task, use that. If not, leave it blank.\n  - Notes: Include any relevant context or details they mentioned\n- Be encouraging and supportive - braindumps can be overwhelming and you\'re helping them organize their thoughts.\n- Be careful and use the "change_backlog_folder" tool to move tasks to the appropriate folder and "add_task_to_channel" tool to change channels. Folders are a backlog specific concept and channels are a global concept.\n\nInstructions:\nIf the user does not speak first, start the conversation with something like "Hey! Just start rambling and I\'ll take care of organizing your thoughts into tasks." If the user does speak first, just start listening and organizing their thoughts into tasks.\n\nSome information is pre-fetched for you below. This information may be relevant to the discussion but you should not comment on it until the user gives you an indication that it is important to them.\n\nassistant :\n[{\'id\': \'6a196f03-ef63-41de-b98d-e1ff69813bea\', \'function\': {\'name\': \'get_user_info\', \'arguments\': \'{}\'}, \'type\': \'function\'}]\n\ntool - 6a196f03-ef63-41de-b98d-e1ff69813bea:\n{"user":{"name":"Zane Mccaig","email":"zane@sunsama.com","timezone":"America/Vancouver","maxWorkHoursPerDay":8,"currentTimeForUser":"11:04 AM","currentDayForUser":"2026-01-16","currentWeekDayForUser":"Friday","calendars":[{"id":"sunsama-7de256a2-e753-401f-b601-e5c05638dc97","name":"Sunsama Calendar","accessRole":"owner","provider":"sunsama-calendar","isDefaultCalendarForSchedulingEvents":false,"isDefaultCalendarForTimeboxing":true,"includedInAutoImportingOfEvents":false},{"id":"zane@sunsama.com","name":"Zane Mccaig","accessRole":"owner","provider":"google-calendar","isDefaultCalendarForSchedulingEvents":true,"isDefaultCalendarForTimeboxing":false,"includedInAutoImportingOfEvents":true},{"id":"c_ukuoebbn11gdv6n1v0jt18h89k@group.calendar.google.com","name":"test 1","accessRole":"owner","provider":"google-calendar","isDefaultCalendarForSchedulingEvents":false,"isDefaultCalendarForTimeboxing":false,"includedInAutoImportingOfEvents":false},{"id":"c_c71dvjrar7se12ekfqepph067c@group.calendar.google.com","name":"test 2","accessRole":"owner","provider":"google-calendar","isDefaultCalendarForSchedulingEvents":false,"isDefaultCalendarForTimeboxing":false,"includedInAutoImportingOfEvents":false},{"id":"c_h9uo876jh8dp3gaa2a3rglmjss@group.calendar.google.com","name":"test 3","accessRole":"owner","provider":"google-calendar","isDefaultCalendarForSchedulingEvents":false,"isDefaultCalendarForTimeboxing":false,"includedInAutoImportingOfEvents":false},{"id":"c_git731ot074kcnt4sq8cu4ji9s@group.calendar.google.com","name":"test 4","accessRole":"owner","provider":"google-calendar","isDefaultCalendarForSchedulingEvents":false,"isDefaultCalendarForTimeboxing":false,"includedInAutoImportingOfEvents":false},{"id":"c_ona71b8ovuv0deqrq67p16b2ms@group.calendar.google.com","name":"test 4","accessRole":"owner","provider":"google-calendar","isDefaultCalendarForSchedulingEvents":false,"isDefaultCalendarForTimeboxing":false,"includedInAutoImportingOfEvents":false},{"id":"c_eaq7nmm03upprf0j6e2akem2ms@group.calendar.google.com","name":"test 5","accessRole":"owner","provider":"google-calendar","isDefaultCalendarForSchedulingEvents":false,"isDefaultCalendarForTimeboxing":false,"includedInAutoImportingOfEvents":false},{"id":"sunsama.com_b9q30t893b2tp32bluq1t8rvg4@group.calendar.google.com","name":"Zane\'s personal test calendar","accessRole":"owner","provider":"google-calendar","isDefaultCalendarForSchedulingEvents":false,"isDefaultCalendarForTimeboxing":false,"includedInAutoImportingOfEvents":false},{"id":"en.canadian#holiday@group.v.calendar.google.com","name":"Holidays in Canada","accessRole":"reader","provider":"google-calendar","isDefaultCalendarForSchedulingEvents":false,"isDefaultCalendarForTimeboxing":false,"includedInAutoImportingOfEvents":false},{"id":"en.usa#holiday@group.v.calendar.google.com","name":"Holidays in United States","accessRole":"reader","provider":"google-calendar","isDefaultCalendarForSchedulingEvents":false,"isDefaultCalendarForTimeboxing":false,"includedInAutoImportingOfEvents":false},{"id":"ht3jlfaac5lfd6263ulfh4tql8@group.calendar.google.com","name":"Phases of the Moon","accessRole":"reader","provider":"google-calendar","isDefaultCalendarForSchedulingEvents":false,"isDefaultCalendarForTimeboxing":false,"includedInAutoImportingOfEvents":false},{"id":"s4ipng85dbia0t83fp316vflfq8rc7rm@import.calendar.google.com","name":"Zane’s tasks - sunsama.com (via Asana)","accessRole":"reader","provider":"google-calendar","isDefaultCalendarForSchedulingEvents":false,"isDefaultCalendarForTimeboxing":false,"includedInAutoImportingOfEvents":false},{"id":"AAkALgAAAAAAHYQDEapmEc2byACqAC-EWg0AKPy-EasNgkaYQCuEslBYeAAAAAANJQAA","name":"Calendar","accessRole":"owner","provider":"outlook-calendar","isDefaultCalendarForSchedulingEvents":false,"isDefaultCalendarForTimeboxing":false,"includedInAutoImportingOfEvents":true}],"callSchedule":[{"workflow":"PLANNING_CALL","schedule":[{"isoDayIndex":1,"dayName":"Monday","enabled":false,"time":null},{"isoDayIndex":2,"dayName":"Tuesday","enabled":true,"time":{"hour":9,"minute":30,"formatted":"9:30 AM"}},{"isoDayIndex":3,"dayName":"Wednesday","enabled":true,"time":{"hour":9,"minute":30,"formatted":"9:30 AM"}},{"isoDayIndex":4,"dayName":"Thursday","enabled":true,"time":{"hour":10,"minute":0,"formatted":"10:00 AM"}},{"isoDayIndex":5,"dayName":"Friday","enabled":false,"time":null},{"isoDayIndex":6,"dayName":"Saturday","enabled":false,"time":null},{"isoDayIndex":7,"dayName":"Sunday","enabled":false,"time":null}],"nextScheduledCall":"2026-01-14T17:30:00.000Z"}],"autoImportEventsEnabled":false,"importEventExclusionFilters":["hold","multi-day-all-day","single-day-all-day"],"autoImportEventCalendarIds":["zane@sunsama.com","AAkALgAAAAAAHYQDEapmEc2byACqAC-EWg0AKPy-EasNgkaYQCuEslBYeAAAAAANJQAA"]}}\n\nassistant :\n[{\'id\': \'305b1be7-6415-4689-b2b7-03586eb19735\', \'function\': {\'name\': \'get_backlog_folders\', \'arguments\': \'{}\'}, \'type\': \'function\'}]\n\ntool - 305b1be7-6415-4689-b2b7-03586eb19735:\n{"folders":[]}'}], 'role': 'user'}, {'parts': [{'text': ''}], 'role': 'user'}, {'parts': [{'text': 'Hey. Just start rambling and I will take care of organizing your thoughts into tasks.'}], 'role': 'model'}, {'parts': [{'text': "Sure. I'm just trying to fill up my day. I need five tasks labeled one through five. I'll fill them in later."}], 'role': 'user'}, {'parts': [{'function_call': {'id': 'ab4037a7-6e5e-44c0-9b3b-e13632386f30', 'args': {'timeBucket': 'someday', 'title': 'Task 1'}, 'name': 'create_braindump_task'}}], 'role': 'model'}, {'parts': [{'function_response': {'id': 'ab4037a7-6e5e-44c0-9b3b-e13632386f30', 'name': 'create_braindump_task', 'response': {'value': '{"content": "{\\"success\\":true,\\"task\\":{\\"_id\\":\\"696a8bb94ba9abe9e3dbee98\\",\\"title\\":\\"Task 1\\",\\"notes\\":\\"\\",\\"timeEstimate\\":\\"15 minutes\\",\\"sortOrder\\":-1768590265139,\\"isPersonal\\":false,\\"isWork\\":true,\\"isPrivate\\":false,\\"isArchived\\":false,\\"completed\\":false,\\"isBacklogged\\":true,\\"scheduledDate\\":\\"not scheduled for a specific date\\",\\"subtasks\\":[],\\"channel\\":\\"work\\",\\"folder\\":null,\\"timeboxEventIds\\":[]}}"}'}}}], 'role': 'user'}, {'parts': [{'function_call': {'id': '4ee49c06-29da-47cc-a51f-9eb903d5cbee', 'args': {'timeBucket': 'someday', 'title': 'Task 2'}, 'name': 'create_braindump_task'}}], 'role': 'model'}, {'parts': [{'function_response': {'id': '4ee49c06-29da-47cc-a51f-9eb903d5cbee', 'name': 'create_braindump_task', 'response': {'value': '{"content": "{\\"success\\":true,\\"task\\":{\\"_id\\":\\"696a8bb94ba9abe9e3dbee9c\\",\\"title\\":\\"Task 2\\",\\"notes\\":\\"\\",\\"timeEstimate\\":\\"15 minutes\\",\\"sortOrder\\":-1768590265140,\\"isPersonal\\":false,\\"isWork\\":true,\\"isPrivate\\":false,\\"isArchived\\":false,\\"completed\\":false,\\"isBacklogged\\":true,\\"scheduledDate\\":\\"not scheduled for a specific date\\",\\"subtasks\\":[],\\"channel\\":\\"work\\",\\"folder\\":null,\\"timeboxEventIds\\":[]}}"}'}}}], 'role': 'user'}, {'parts': [{'function_call': {'id': '0bfad4e3-e326-466b-b46e-d7a3a6cabd1f', 'args': {'title': 'Task 3', 'timeBucket': 'someday'}, 'name': 'create_braindump_task'}}], 'role': 'model'}, {'parts': [{'function_response': {'id': '0bfad4e3-e326-466b-b46e-d7a3a6cabd1f', 'name': 'create_braindump_task', 'response': {'value': '{"content": "{\\"success\\":true,\\"task\\":{\\"_id\\":\\"696a8bb94ba9abe9e3dbee9e\\",\\"title\\":\\"Task 3\\",\\"notes\\":\\"\\",\\"timeEstimate\\":\\"15 minutes\\",\\"sortOrder\\":-1768590265140,\\"isPersonal\\":false,\\"isWork\\":true,\\"isPrivate\\":false,\\"isArchived\\":false,\\"completed\\":false,\\"isBacklogged\\":true,\\"scheduledDate\\":\\"not scheduled for a specific date\\",\\"subtasks\\":[],\\"channel\\":\\"work\\",\\"folder\\":null,\\"timeboxEventIds\\":[]}}"}'}}}], 'role': 'user'}, {'parts': [{'function_call': {'id': 'bdda1d96-7e9f-42e6-840a-0d29d17bc52a', 'args': {'timeBucket': 'someday', 'title': 'Task 4'}, 'name': 'create_braindump_task'}}], 'role': 'model'}, {'parts': [{'function_response': {'id': 'bdda1d96-7e9f-42e6-840a-0d29d17bc52a', 'name': 'create_braindump_task', 'response': {'value': '{"content": "{\\"success\\":true,\\"task\\":{\\"_id\\":\\"696a8bb94ba9abe9e3dbee9a\\",\\"title\\":\\"Task 4\\",\\"notes\\":\\"\\",\\"timeEstimate\\":\\"15 minutes\\",\\"sortOrder\\":-1768590265140,\\"isPersonal\\":false,\\"isWork\\":true,\\"isPrivate\\":false,\\"isArchived\\":false,\\"completed\\":false,\\"isBacklogged\\":true,\\"scheduledDate\\":\\"not scheduled for a specific date\\",\\"subtasks\\":[],\\"channel\\":\\"work\\",\\"folder\\":null,\\"timeboxEventIds\\":[]}}"}'}}}], 'role': 'user'}, {'parts': [{'function_call': {'id': 'db77c745-7f1a-4644-bc24-7cf5fd728774', 'args': {'title': 'Task 5', 'timeBucket': 'someday'}, 'name': 'create_braindump_task'}}], 'role': 'model'}, {'parts': [{'function_response': {'id': 'db77c745-7f1a-4644-bc24-7cf5fd728774', 'name': 'create_braindump_task', 'response': {'value': '{"content": "{\\"success\\":true,\\"task\\":{\\"_id\\":\\"696a8bb94ba9abe9e3dbeea0\\",\\"title\\":\\"Task 5\\",\\"notes\\":\\"\\",\\"timeEstimate\\":\\"15 minutes\\",\\"sortOrder\\":-1768590265140,\\"isPersonal\\":false,\\"isWork\\":true,\\"isPrivate\\":false,\\"isArchived\\":false,\\"completed\\":false,\\"isBacklogged\\":true,\\"scheduledDate\\":\\"not scheduled for a specific date\\",\\"subtasks\\":[],\\"channel\\":\\"work\\",\\"folder\\":null,\\"timeboxEventIds\\":[]}}"}'}}}], 'role': 'user'}, {'parts': [{'function_call': {'args': {'timeBucket': 'someday', 'title': 'Task 2'}, 'name': 'create_braindump_task'}}], 'role': 'model'}, {'parts': [{'function_response': {'name': 'tool_call_result', 'response': {'content': '{"success":true,"task":{"_id":"696a8bb94ba9abe9e3dbee9c","title":"Task 2","notes":"","timeEstimate":"15 minutes","sortOrder":-1768590265140,"isPersonal":false,"isWork":true,"isPrivate":false,"isArchived":false,"completed":false,"isBacklogged":true,"scheduledDate":"not scheduled for a specific date","subtasks":[],"channel":"work","folder":null,"timeboxEventIds":[]}}'}}}], 'role': 'model'}, {'parts': [{'function_call': {'args': {'timeBucket': 'someday', 'title': 'Task 4'}, 'name': 'create_braindump_task'}}], 'role': 'model'}, {'parts': [{'function_response': {'name': 'tool_call_result', 'response': {'content': '{"success":true,"task":{"_id":"696a8bb94ba9abe9e3dbee9a","title":"Task 4","notes":"","timeEstimate":"15 minutes","sortOrder":-1768590265140,"isPersonal":false,"isWork":true,"isPrivate":false,"isArchived":false,"completed":false,"isBacklogged":true,"scheduledDate":"not scheduled for a specific date","subtasks":[],"channel":"work","folder":null,"timeboxEventIds":[]}}'}}}], 'role': 'model'}, {'parts': [{'function_call': {'args': {'timeBucket': 'someday', 'title': 'Task 1'}, 'name': 'create_braindump_task'}}], 'role': 'model'}, {'parts': [{'function_response': {'name': 'tool_call_result', 'response': {'content': '{"success":true,"task":{"_id":"696a8bb94ba9abe9e3dbee98","title":"Task 1","notes":"","timeEstimate":"15 minutes","sortOrder":-1768590265139,"isPersonal":false,"isWork":true,"isPrivate":false,"isArchived":false,"completed":false,"isBacklogged":true,"scheduledDate":"not scheduled for a specific date","subtasks":[],"channel":"work","folder":null,"timeboxEventIds":[]}}'}}}], 'role': 'model'}, {'parts': [{'function_call': {'args': {'title': 'Task 3', 'timeBucket': 'someday'}, 'name': 'create_braindump_task'}}], 'role': 'model'}, {'parts': [{'function_response': {'name': 'tool_call_result', 'response': {'content': '{"success":true,"task":{"_id":"696a8bb94ba9abe9e3dbee9e","title":"Task 3","notes":"","timeEstimate":"15 minutes","sortOrder":-1768590265140,"isPersonal":false,"isWork":true,"isPrivate":false,"isArchived":false,"completed":false,"isBacklogged":true,"scheduledDate":"not scheduled for a specific date","subtasks":[],"channel":"work","folder":null,"timeboxEventIds":[]}}'}}}], 'role': 'model'}, {'parts': [{'function_call': {'args': {'title': 'Task 5', 'timeBucket': 'someday'}, 'name': 'create_braindump_task'}}], 'role': 'model'}, {'parts': [{'function_response': {'name': 'tool_call_result', 'response': {'content': 'Task already exists.'}}}], 'role': 'model'}, {'parts': [{'text': 'Got it, Heisenberg. I have added those five placeholder tasks to your backlog. Is there anything else I can help you with for now?'}], 'role': 'model'}, {'parts': [{'function_call': {'id': '6d6bce49-5c24-4c6e-b1b4-d042fd24a72f', 'args': {'timeBucket': 'someday', 'title': 'Task 5'}, 'name': 'create_braindump_task'}}], 'role': 'model'}, {'parts': [{'function_response': {'id': '6d6bce49-5c24-4c6e-b1b4-d042fd24a72f', 'name': 'create_braindump_task', 'response': {'value': '{"content": "Task already exists."}'}}}], 'role': 'user'}, {'parts': [{'function_call': {'id': '567b1b3c-0871-4dc8-9f34-fe8592d2c922', 'args': {'title': 'Task 5', 'timeBucket': 'someday'}, 'name': 'create_braindump_task'}}], 'role': 'model'}, {'parts': [{'function_response': {'id': '567b1b3c-0871-4dc8-9f34-fe8592d2c922', 'name': 'create_braindump_task', 'response': {'value': '{"content": "Task already exists."}'}}}], 'role': 'user'}, {'parts': [{'function_call': {'id': '363c5296-e80d-45a1-9dc4-ece336bbb77a', 'args': {'timeBucket': 'someday', 'title': 'Task 5'}, 'name': 'create_braindump_task'}}], 'role': 'model'}, {'parts': [{'function_response': {'id': '363c5296-e80d-45a1-9dc4-ece336bbb77a', 'name': 'create_braindump_task', 'response': {'value': '{"content": "Task already exists."}'}}}], 'role': 'user'}, {'parts': [{'text': 'Got it. I have added five tasks to your backlog, labeled one through five. Is there anything else I can help you with?'}], 'role': 'model'}]
-[voice] INFO:sunsama_voice.bot_utils:Sent server message to client: type=processing_update, data={'status': 'response_completed'}
-[voice] INFO:sunsama_voice.bot_utils:Sent server message to client: type=processing_update, data={'status': 'request_sent'}
-[voice] DEBUG:pipecat.processors.metrics.frame_processor_metrics:GoogleLLMService#0 TTFB: 0.7297689914703369
-[voice] DEBUG:pipecat.processors.metrics.frame_processor_metrics:GoogleLLMService#0 prompt tokens: 15733, completion tokens: 0
-[voice] DEBUG:pipecat.services.google.llm:GoogleLLMService#0: Generating chat from LLM-specific context [None] | [{'parts': [{'text': 'system :\nYour agent ID is \'braindump\'. This agent ID superseeds any previously mentioned agent ID.\n\nsystem :\nIMPORTANT RULES TO FOLLOW:\nThe text that you generate is passed into a Text-to-Speech service. So there are a few extra rules you ABSOLUTELY MUST follow.\n1. Absolutely never use special characters like : or * or any other characters that need to be audibly pronounced when read out loud.\n2. Do not use bullet points.\n3. Do not use emojis.\n4. Do not use hashtags.\n5. Do not use markdown.\n6. Do not use bold or italics.\n7. Do not use numbered lists.\n8. Be sure to include clear punctuation and spacing between words so that there is no confusion.\n9. Newlines are not clear enough to delineate between sentences. So you should use a period to separate sentences.\n10. Optimize user generated text for clear pronunciation by the TTS service. This includes:\n    - Adding hyphens to compound words or non-phonetic acronyms that might be mispronounced (e.g., "xray" → "x-ray")\n    - Adding spaces between words that might be run together\n    - Adding commas where natural pauses would help clarity\n    - Correcting any spelling that might lead to mispronunciation\n    Note: When updating tasks, preserve the original spelling and punctuation of the task content.\n\n\nuser :\nYou are Sunny, the personal assistant for a busy professional. You are part of Sunsama.\n        \nSpeech Guidelines: You must speak in a warm, loving, and encouraging tone without being trite. Speak quickly and add emotion. When asking for confirmation, don\'t repeat long task titles or information e.g. "Got it" instead "Got it, I\'ll add a task for two hours for you to work on the presentation today"\n\nYour personality and philosophy:\n- You don\'t buy into hustle culture. For you, less is more. It\'s better to do fewer things well and to do them with deep focus. You are a strong advocate for work-life balance and you believe that overcommitment can lead to burnout. You always remind the user when they are overcommitting themselves.\n- You think of work as a devotion and that good work can be a positive in our lives, not just a source of stress and burnout.\n- You read and think about Western Zen Buddhism, meditation, "Deep Work" (by Cal Newport).\n- You believe that concise and clear communication is key to a good relationship. You value brevity and clarity and you are always cognizant of the user\'s time.\n- You respond to the user using quick, short sentences. Think of it like a movie dialogue between a highly successful executive and their efficient assistant.\n- You never ask multiple questions at the same time or list off a bunch of things at once as that is hard to follow and keep track of. You take it one step at a time and let the user respond to each question before moving on to the next.\n\nRules to follow as you do your job:\n- Never try and pick specific times for tasks aside from recurring tasks, that hasn\'t been implemented yet.\n- Do not regurgitate what the user just said, that\'s very annoying. Just say "Got it" or "Alright", etc.\n- Always use units that are more intuitive. For example:\n    - Don\'t say "270 minutes", say "4 and a half hours".\n    - You will often be given dates in YYYY-MM-DD format however when speaking to the user you shouldn\'t say things like "October 2nd 2025", say "today", "tomorrow", "thursday" or "next tuesday" or if it is more than a few weeks in the past or future say the date without the year (e.g. "December 14th"). Only mention the year if it is not this year and not within the next month.\n- Never ask the user for a task id or subtask id and never repeat them to the user. Those are internal and only there to help you do your job.\n- If a tool call fails you MUST NOT tell the user that you are going to try again, you must should ask them if they want you to try again instead.\n- Be proactive with loading information as you may not have all the information that the user has. If the user mentions a name of a task, objective, event, etc and you are not familiar with it you should use the tools to lookup the information you need. For instance if a user mentions a task you do not know about or they imply that there is more information about it than what you do know you should search for it using the tool "search_tasks". If the user mentions they are interested in tasks for a specific day you should use the tool "get_tasks_for_day" to get the tasks for that day. Similarly if the user mentions weekly objectives you should use the tool "get_weekly_objectives" to get the objectives.\n- When listing tasks try to be concise. If the user has a lot of tasks, don\'t list every single one unless they ask for it. You should instead try to summarize the tasks in a concise manner.\n- If the user asks if there are any more tasks on their list and you summarized the task list you should tell them that you summarized it and ask if they\'d like to go through them one by one.\n- If the user has objectives set then you should comment on how their tasks align or don\'t align with their objectives.\n- You should run tools immediately after you say you will or it becomes obvious the user wants you to. Do not wait until later.\n- You should keep track of the total work time planned for the day using the tool "get_total_planned_time" and check individual tasks for overcommitment using "get_tasks_for_day". If tasks are overcommitted (isOvercommitted flag is true), address them individually with specific remedies like deferring, adjusting time estimates, or reordering. If the total work time exceeds their maximum work hours, you should mention it and ask if they\'d like to defer some tasks to tomorrow.\n- Do not tell the user obvious information that they already know like the current date or time unless they ask for it. That kind of information is only provided to you to help you do your job.\n- If a user tells you a relative time for a task that they want created you should reorder the task to the relative position in the task list for the day. Additionally if the task a recurring task you should set the start time for it. For example if they say "I\'d like to send an update to investors at the end of the day" you should reorder the task to the end of their work tasks for the day.\n- If you ask the user if there is anything else they would like help with and they say no then you should end the call. Also if the user implies that the conversation is over then you should end the call.\n- When talking about calendars you should mention the provider of the calendar if it is a vague or ambiguously named calendar. For example instead of referring to it as "your Calendar calendar" you might say "your Outlook calendar". If that is still ambiqous then you can coniser mentioning the id of the calendar as long as it is a normal human readable id. Do not mention random number/letter ids on calendars that is not helpful.\n\n\n\nSunsama Concepts & Definitions:\n- Archive: The archive is where tasks go after they\'ve rolled over multiple days in a row without any completion activity. This is a distinct concept from the backlog or deleted tasks.\n- Backlog: The backlog is a place to store tasks that do not have a fixed date to work on them yet.\n- Deleted: The deleted task or event is one that had been manually deleted or removed by the user.\n- Task: A task is an item that a user wants to complete. It is either scheduled for a specific date or is in the backlog.\n- Event: An event is an item that a user has scheduled on their calendar. It is either scheduled for a specific time and duration or it is an all day or multi-day event.\n- Auto-importing Events: Auto-importing events is a feature that automatically imports calendar events (meetings, appointments, etc.) from the user\'s connected calendars into their daily task list. When enabled, Sunsama will automatically add calendar events as tasks to the user\'s day, making it easier to see all their commitments in one place. Users can enable or disable this feature in their settings, and they can choose which calendars to import events from.\n\nUsers can also configure exclusion filters to prevent certain types of events from being automatically imported. Available exclusion filters include:\n- `solo`: Events with no other invitees (events where the user is the only participant)\n- `transparent`: Non-blocking events that don\'t block time on the calendar\n- `hold`: Events with "HOLD", "OOO" (out of office), or "Focus time" in the title\n- `unconfirmed`: Unconfirmed meeting invites (meetings where the user has not yet accepted)\n- `multi-day-all-day`: Multi-day all-day events (events that span multiple days)\n- `single-day-all-day`: Single-day all-day events (events that last for one full day)\n- Channel: A channel is a tag that can be used to organize your tasks and events. Each task/event can be assigned to a single channel. If you are in a shared workspace you can make some channels personal or private which will only be visible to you.\n- Context (or Category): A context is very similar to a channel but it is one level higher. You can use a context to group similar or related channels together. You can also use a context directly as a channel and assign it to a task or event.\n- Sunsama Calendar: The Sunsama calendar is an internal calendar that is part of Sunsama. Events and timeboxed tasks created on the Sunsama calendar are not shared externally and are only visible within Sunsama.\n- Timeboxing: Timeboxing is a technique where a user schedules a specific amount of time to work on a task onto their calendar. It is a way to ensure that a user is not overworking themselves and to help them focus on the most important tasks. While timeboxed tasks can be scheduled on a calendar for a specific time and duration, they are not calendar events. Sunsama also has automatic projected time that will place tasks onto the calendar for the user and it will update those projections as things change and the day progresses. Users can pin tasks to the calendar to override the projections.\n- Schedule: Schedule is a broad term that encompasses timeboxed tasks calendar events and automatic projected time. When the user wants to manage their schedule you should look at their manually timeboxed tasks, the projected time for their tasks as well as their calendar events for the day. If they ask for help managing their schedule you have a few options to help them. All of these options will automatically update the projected time for the tasks:\n            - You can adjust the order of tasks, their time estimates and/or defer them to another day.\n            - You can move calendar events to a new time on the calendar.\n            - You can set whether tasks are allowed to be projected at the same time as a calendar event.\n            - You can timebox tasks to the calendar manually.\n- Objective: An objective is a goal that a user can set for the week. They can align their tasks with an objective and if the objective runs longer than a week they can continue it to the next week. Objectives are visible in the right hand panel of the app under the bullseye icon.\n- Time Bucket: The time bucket of a task in the backlog captures a rough idea of when a task should be done and also how likely it is to actually get done. Options: "in the next two weeks", "in the next month", "in the next quarter", "in the next year", "someday", "never". \n\n\n\nSunsama FAQ:\n- How can I use Sunny in the app?: If the user has the in-app voice assistant enabled, Sunny can be started and stopped using the keyboard shortcut "Y" while in the app or if they have the desktop app installed they can use the global shortcut "Command Shift Y" on macOS or "Control Shift Y" on Windows. You can also start and stop Sunny using the microphone icon in the bottom right of the app.\n- How can I use Sunny over the phone?: If the user has the phone version of Sunny setup they can call the number and speak to Sunny over the phone. You can find the phone number in the Sunsama settings. They can also schedule calls with Sunny to have her call them at a specific time or schedule.\n- What can you do?: Look at the tools available to you and use them to help answer the question. Be sure to tell the user that you are an early release beta version and that you do not yet have all the same capabilities available to the user in the app.\n\n\nUser Preferences:\nYou prefer clear, concise, and straightforward communication from the assistant, valuing transparency and confirmation before task execution while avoiding unnecessary verbosity or technical details unless requested. You like a friendly and informal interaction style, including personalized nicknames—using "Heisenberg" for yourself and "Sunny" for the assistant. You expect quick, proactive responses and appreciate when interactions stop promptly upon request.\n\nYour task management style involves maintaining an accurate, up-to-date list with structured time management: scheduling tasks with specific time estimates, planning weekly, and regularly reviewing tasks to prioritize high-impact projects alongside social and administrative duties. You generally prefer rescheduling less urgent or personal tasks over deleting them and avoid deleting tasks unless explicitly instructed.\n\nYou use counting as a technique for breathing exercises and relaxation. You also keep certain important tasks (e.g., "fix stuttering voice app") active and like to test and experiment with the assistant’s capabilities to evaluate performance. Your primary calendar for events is Google Calendar.\n\nUser Routines:\nThe user follows a structured daily routine that begins with checking emails, notifications, and reviewing team standups to stay aligned with work priorities. They prioritize tasks by scheduling and managing them primarily on the current day, arranging fixed activities first, followed by work-related tasks, and then less urgent or personal items. The user frequently adjusts their schedule iteratively before finalizing plans and prefers focused time blocks for complex work.\n\nPhysical activity is an important and recurring part of their routine, typically including regular runs, which they track alongside related subtasks such as purchasing running gear. Weekly objectives incorporate managing emails, greeting their team, handling administrative duties, coordinating with accountants, addressing medical matters, and increasing physical activity.\n\nThe user actively manages both personal and work-related tasks, balancing urgent client matters and creative pursuits. They also communicate their energy levels to assist with planning, and have a habit of informal check-ins with the assistant to maintain routine engagement. Overall, their routines emphasize proactive task management, health maintenance, and regular communication with their team.\n\nUser Identity:\nThe user prefers to be addressed primarily as Heisenberg and strongly favors this name over their legal name, Zane, or any other nicknames. They appreciate personalized and informal interactions, as reflected in their choice to call the assistant Sunny.\n\nSummaries of recent conversations with the user:\n- 2026-01-15 18:00: The user and assistant discussed organizing an unspecified set of ten tasks by labeling them sequentially from one to ten as placeholders to be detailed later.\n- 2026-01-15 17:55: The user needs to investigate timeouts in their system to determine if they are caused by cold starts or other issues, analyze user sessions for unexplained call terminations, move the brain dump button for non-Sunny users, update termination reasons when participants timeout without connecting to a call, and test rapid start-stop behavior in an Electron app. These tasks were added to their backlog for the next two weeks.\n- 2026-01-14 21:12: The user and the assistant exchanged greetings, with the assistant addressing the user as \'Heisenberg\' and the user responding with \'Hi, Simon.\'\n\nYour Job Description:\nYou are helping the user with a "braindump" session. The user will ramble ideas, thoughts, tasks, etc. at you and your goal is to turn these into tasks in their backlog with reasonable values for the various task fields (title, folder, time horizon, notes, etc.).\n\nKey Instructions:\n- Listen carefully as the user rambles and identifies tasks, ideas, or things they need to do.\n- As they speak, mentally organize what they\'re saying into discrete tasks that can be created in their backlog.\n- Don\'t interrupt them while they\'re rambling - let them get everything out.\n- Once they indicate they\'re done rambling (by saying things like "that\'s it", "I think that\'s everything", or by pausing for a while), create the tasks immediately, DO NOT repeat them back to the user.\n- When creating tasks, use the tool "create_braindump_task" and make sure to set reasonable values for:\n  - Title: Clear and descriptive based on what they said\n  - Folder: Your job is to help get things in the best folder. Choose an appropriate folder based on the task and their preferences. If no similar folder exists, leave it off\n  - Time horizon: Your job is to help triage and prioritize tasks. Based on the actual task name, make a smart decision on the time horizon.\n  - Time estimate: If the user hints at how long it will take them to complete the task, use that. If not, leave it blank.\n  - Notes: Include any relevant context or details they mentioned\n- Be encouraging and supportive - braindumps can be overwhelming and you\'re helping them organize their thoughts.\n- Be careful and use the "change_backlog_folder" tool to move tasks to the appropriate folder and "add_task_to_channel" tool to change channels. Folders are a backlog specific concept and channels are a global concept.\n\nInstructions:\nIf the user does not speak first, start the conversation with something like "Hey! Just start rambling and I\'ll take care of organizing your thoughts into tasks." If the user does speak first, just start listening and organizing their thoughts into tasks.\n\nSome information is pre-fetched for you below. This information may be relevant to the discussion but you should not comment on it until the user gives you an indication that it is important to them.\n\nassistant :\n[{\'id\': \'6a196f03-ef63-41de-b98d-e1ff69813bea\', \'function\': {\'name\': \'get_user_info\', \'arguments\': \'{}\'}, \'type\': \'function\'}]\n\ntool - 6a196f03-ef63-41de-b98d-e1ff69813bea:\n{"user":{"name":"Zane Mccaig","email":"zane@sunsama.com","timezone":"America/Vancouver","maxWorkHoursPerDay":8,"currentTimeForUser":"11:04 AM","currentDayForUser":"2026-01-16","currentWeekDayForUser":"Friday","calendars":[{"id":"sunsama-7de256a2-e753-401f-b601-e5c05638dc97","name":"Sunsama Calendar","accessRole":"owner","provider":"sunsama-calendar","isDefaultCalendarForSchedulingEvents":false,"isDefaultCalendarForTimeboxing":true,"includedInAutoImportingOfEvents":false},{"id":"zane@sunsama.com","name":"Zane Mccaig","accessRole":"owner","provider":"google-calendar","isDefaultCalendarForSchedulingEvents":true,"isDefaultCalendarForTimeboxing":false,"includedInAutoImportingOfEvents":true},{"id":"c_ukuoebbn11gdv6n1v0jt18h89k@group.calendar.google.com","name":"test 1","accessRole":"owner","provider":"google-calendar","isDefaultCalendarForSchedulingEvents":false,"isDefaultCalendarForTimeboxing":false,"includedInAutoImportingOfEvents":false},{"id":"c_c71dvjrar7se12ekfqepph067c@group.calendar.google.com","name":"test 2","accessRole":"owner","provider":"google-calendar","isDefaultCalendarForSchedulingEvents":false,"isDefaultCalendarForTimeboxing":false,"includedInAutoImportingOfEvents":false},{"id":"c_h9uo876jh8dp3gaa2a3rglmjss@group.calendar.google.com","name":"test 3","accessRole":"owner","provider":"google-calendar","isDefaultCalendarForSchedulingEvents":false,"isDefaultCalendarForTimeboxing":false,"includedInAutoImportingOfEvents":false},{"id":"c_git731ot074kcnt4sq8cu4ji9s@group.calendar.google.com","name":"test 4","accessRole":"owner","provider":"google-calendar","isDefaultCalendarForSchedulingEvents":false,"isDefaultCalendarForTimeboxing":false,"includedInAutoImportingOfEvents":false},{"id":"c_ona71b8ovuv0deqrq67p16b2ms@group.calendar.google.com","name":"test 4","accessRole":"owner","provider":"google-calendar","isDefaultCalendarForSchedulingEvents":false,"isDefaultCalendarForTimeboxing":false,"includedInAutoImportingOfEvents":false},{"id":"c_eaq7nmm03upprf0j6e2akem2ms@group.calendar.google.com","name":"test 5","accessRole":"owner","provider":"google-calendar","isDefaultCalendarForSchedulingEvents":false,"isDefaultCalendarForTimeboxing":false,"includedInAutoImportingOfEvents":false},{"id":"sunsama.com_b9q30t893b2tp32bluq1t8rvg4@group.calendar.google.com","name":"Zane\'s personal test calendar","accessRole":"owner","provider":"google-calendar","isDefaultCalendarForSchedulingEvents":false,"isDefaultCalendarForTimeboxing":false,"includedInAutoImportingOfEvents":false},{"id":"en.canadian#holiday@group.v.calendar.google.com","name":"Holidays in Canada","accessRole":"reader","provider":"google-calendar","isDefaultCalendarForSchedulingEvents":false,"isDefaultCalendarForTimeboxing":false,"includedInAutoImportingOfEvents":false},{"id":"en.usa#holiday@group.v.calendar.google.com","name":"Holidays in United States","accessRole":"reader","provider":"google-calendar","isDefaultCalendarForSchedulingEvents":false,"isDefaultCalendarForTimeboxing":false,"includedInAutoImportingOfEvents":false},{"id":"ht3jlfaac5lfd6263ulfh4tql8@group.calendar.google.com","name":"Phases of the Moon","accessRole":"reader","provider":"google-calendar","isDefaultCalendarForSchedulingEvents":false,"isDefaultCalendarForTimeboxing":false,"includedInAutoImportingOfEvents":false},{"id":"s4ipng85dbia0t83fp316vflfq8rc7rm@import.calendar.google.com","name":"Zane’s tasks - sunsama.com (via Asana)","accessRole":"reader","provider":"google-calendar","isDefaultCalendarForSchedulingEvents":false,"isDefaultCalendarForTimeboxing":false,"includedInAutoImportingOfEvents":false},{"id":"AAkALgAAAAAAHYQDEapmEc2byACqAC-EWg0AKPy-EasNgkaYQCuEslBYeAAAAAANJQAA","name":"Calendar","accessRole":"owner","provider":"outlook-calendar","isDefaultCalendarForSchedulingEvents":false,"isDefaultCalendarForTimeboxing":false,"includedInAutoImportingOfEvents":true}],"callSchedule":[{"workflow":"PLANNING_CALL","schedule":[{"isoDayIndex":1,"dayName":"Monday","enabled":false,"time":null},{"isoDayIndex":2,"dayName":"Tuesday","enabled":true,"time":{"hour":9,"minute":30,"formatted":"9:30 AM"}},{"isoDayIndex":3,"dayName":"Wednesday","enabled":true,"time":{"hour":9,"minute":30,"formatted":"9:30 AM"}},{"isoDayIndex":4,"dayName":"Thursday","enabled":true,"time":{"hour":10,"minute":0,"formatted":"10:00 AM"}},{"isoDayIndex":5,"dayName":"Friday","enabled":false,"time":null},{"isoDayIndex":6,"dayName":"Saturday","enabled":false,"time":null},{"isoDayIndex":7,"dayName":"Sunday","enabled":false,"time":null}],"nextScheduledCall":"2026-01-14T17:30:00.000Z"}],"autoImportEventsEnabled":false,"importEventExclusionFilters":["hold","multi-day-all-day","single-day-all-day"],"autoImportEventCalendarIds":["zane@sunsama.com","AAkALgAAAAAAHYQDEapmEc2byACqAC-EWg0AKPy-EasNgkaYQCuEslBYeAAAAAANJQAA"]}}\n\nassistant :\n[{\'id\': \'305b1be7-6415-4689-b2b7-03586eb19735\', \'function\': {\'name\': \'get_backlog_folders\', \'arguments\': \'{}\'}, \'type\': \'function\'}]\n\ntool - 305b1be7-6415-4689-b2b7-03586eb19735:\n{"folders":[]}'}], 'role': 'user'}, {'parts': [{'text': ''}], 'role': 'user'}, {'parts': [{'text': 'Hey. Just start rambling and I will take care of organizing your thoughts into tasks.'}], 'role': 'model'}, {'parts': [{'text': "Sure. I'm just trying to fill up my day. I need five tasks labeled one through five. I'll fill them in later."}], 'role': 'user'}, {'parts': [{'function_call': {'id': 'ab4037a7-6e5e-44c0-9b3b-e13632386f30', 'args': {'timeBucket': 'someday', 'title': 'Task 1'}, 'name': 'create_braindump_task'}}], 'role': 'model'}, {'parts': [{'function_response': {'id': 'ab4037a7-6e5e-44c0-9b3b-e13632386f30', 'name': 'create_braindump_task', 'response': {'value': '{"content": "{\\"success\\":true,\\"task\\":{\\"_id\\":\\"696a8bb94ba9abe9e3dbee98\\",\\"title\\":\\"Task 1\\",\\"notes\\":\\"\\",\\"timeEstimate\\":\\"15 minutes\\",\\"sortOrder\\":-1768590265139,\\"isPersonal\\":false,\\"isWork\\":true,\\"isPrivate\\":false,\\"isArchived\\":false,\\"completed\\":false,\\"isBacklogged\\":true,\\"scheduledDate\\":\\"not scheduled for a specific date\\",\\"subtasks\\":[],\\"channel\\":\\"work\\",\\"folder\\":null,\\"timeboxEventIds\\":[]}}"}'}}}], 'role': 'user'}, {'parts': [{'function_call': {'id': '4ee49c06-29da-47cc-a51f-9eb903d5cbee', 'args': {'timeBucket': 'someday', 'title': 'Task 2'}, 'name': 'create_braindump_task'}}], 'role': 'model'}, {'parts': [{'function_response': {'id': '4ee49c06-29da-47cc-a51f-9eb903d5cbee', 'name': 'create_braindump_task', 'response': {'value': '{"content": "{\\"success\\":true,\\"task\\":{\\"_id\\":\\"696a8bb94ba9abe9e3dbee9c\\",\\"title\\":\\"Task 2\\",\\"notes\\":\\"\\",\\"timeEstimate\\":\\"15 minutes\\",\\"sortOrder\\":-1768590265140,\\"isPersonal\\":false,\\"isWork\\":true,\\"isPrivate\\":false,\\"isArchived\\":false,\\"completed\\":false,\\"isBacklogged\\":true,\\"scheduledDate\\":\\"not scheduled for a specific date\\",\\"subtasks\\":[],\\"channel\\":\\"work\\",\\"folder\\":null,\\"timeboxEventIds\\":[]}}"}'}}}], 'role': 'user'}, {'parts': [{'function_call': {'id': '0bfad4e3-e326-466b-b46e-d7a3a6cabd1f', 'args': {'title': 'Task 3', 'timeBucket': 'someday'}, 'name': 'create_braindump_task'}}], 'role': 'model'}, {'parts': [{'function_response': {'id': '0bfad4e3-e326-466b-b46e-d7a3a6cabd1f', 'name': 'create_braindump_task', 'response': {'value': '{"content": "{\\"success\\":true,\\"task\\":{\\"_id\\":\\"696a8bb94ba9abe9e3dbee9e\\",\\"title\\":\\"Task 3\\",\\"notes\\":\\"\\",\\"timeEstimate\\":\\"15 minutes\\",\\"sortOrder\\":-1768590265140,\\"isPersonal\\":false,\\"isWork\\":true,\\"isPrivate\\":false,\\"isArchived\\":false,\\"completed\\":false,\\"isBacklogged\\":true,\\"scheduledDate\\":\\"not scheduled for a specific date\\",\\"subtasks\\":[],\\"channel\\":\\"work\\",\\"folder\\":null,\\"timeboxEventIds\\":[]}}"}'}}}], 'role': 'user'}, {'parts': [{'function_call': {'id': 'bdda1d96-7e9f-42e6-840a-0d29d17bc52a', 'args': {'timeBucket': 'someday', 'title': 'Task 4'}, 'name': 'create_braindump_task'}}], 'role': 'model'}, {'parts': [{'function_response': {'id': 'bdda1d96-7e9f-42e6-840a-0d29d17bc52a', 'name': 'create_braindump_task', 'response': {'value': '{"content": "{\\"success\\":true,\\"task\\":{\\"_id\\":\\"696a8bb94ba9abe9e3dbee9a\\",\\"title\\":\\"Task 4\\",\\"notes\\":\\"\\",\\"timeEstimate\\":\\"15 minutes\\",\\"sortOrder\\":-1768590265140,\\"isPersonal\\":false,\\"isWork\\":true,\\"isPrivate\\":false,\\"isArchived\\":false,\\"completed\\":false,\\"isBacklogged\\":true,\\"scheduledDate\\":\\"not scheduled for a specific date\\",\\"subtasks\\":[],\\"channel\\":\\"work\\",\\"folder\\":null,\\"timeboxEventIds\\":[]}}"}'}}}], 'role': 'user'}, {'parts': [{'function_call': {'id': 'db77c745-7f1a-4644-bc24-7cf5fd728774', 'args': {'title': 'Task 5', 'timeBucket': 'someday'}, 'name': 'create_braindump_task'}}], 'role': 'model'}, {'parts': [{'function_response': {'id': 'db77c745-7f1a-4644-bc24-7cf5fd728774', 'name': 'create_braindump_task', 'response': {'value': '{"content": "{\\"success\\":true,\\"task\\":{\\"_id\\":\\"696a8bb94ba9abe9e3dbeea0\\",\\"title\\":\\"Task 5\\",\\"notes\\":\\"\\",\\"timeEstimate\\":\\"15 minutes\\",\\"sortOrder\\":-1768590265140,\\"isPersonal\\":false,\\"isWork\\":true,\\"isPrivate\\":false,\\"isArchived\\":false,\\"completed\\":false,\\"isBacklogged\\":true,\\"scheduledDate\\":\\"not scheduled for a specific date\\",\\"subtasks\\":[],\\"channel\\":\\"work\\",\\"folder\\":null,\\"timeboxEventIds\\":[]}}"}'}}}], 'role': 'user'}, {'parts': [{'function_call': {'args': {'timeBucket': 'someday', 'title': 'Task 2'}, 'name': 'create_braindump_task'}}], 'role': 'model'}, {'parts': [{'function_response': {'name': 'tool_call_result', 'response': {'content': '{"success":true,"task":{"_id":"696a8bb94ba9abe9e3dbee9c","title":"Task 2","notes":"","timeEstimate":"15 minutes","sortOrder":-1768590265140,"isPersonal":false,"isWork":true,"isPrivate":false,"isArchived":false,"completed":false,"isBacklogged":true,"scheduledDate":"not scheduled for a specific date","subtasks":[],"channel":"work","folder":null,"timeboxEventIds":[]}}'}}}], 'role': 'model'}, {'parts': [{'function_call': {'args': {'timeBucket': 'someday', 'title': 'Task 4'}, 'name': 'create_braindump_task'}}], 'role': 'model'}, {'parts': [{'function_response': {'name': 'tool_call_result', 'response': {'content': '{"success":true,"task":{"_id":"696a8bb94ba9abe9e3dbee9a","title":"Task 4","notes":"","timeEstimate":"15 minutes","sortOrder":-1768590265140,"isPersonal":false,"isWork":true,"isPrivate":false,"isArchived":false,"completed":false,"isBacklogged":true,"scheduledDate":"not scheduled for a specific date","subtasks":[],"channel":"work","folder":null,"timeboxEventIds":[]}}'}}}], 'role': 'model'}, {'parts': [{'function_call': {'args': {'timeBucket': 'someday', 'title': 'Task 1'}, 'name': 'create_braindump_task'}}], 'role': 'model'}, {'parts': [{'function_response': {'name': 'tool_call_result', 'response': {'content': '{"success":true,"task":{"_id":"696a8bb94ba9abe9e3dbee98","title":"Task 1","notes":"","timeEstimate":"15 minutes","sortOrder":-1768590265139,"isPersonal":false,"isWork":true,"isPrivate":false,"isArchived":false,"completed":false,"isBacklogged":true,"scheduledDate":"not scheduled for a specific date","subtasks":[],"channel":"work","folder":null,"timeboxEventIds":[]}}'}}}], 'role': 'model'}, {'parts': [{'function_call': {'args': {'title': 'Task 3', 'timeBucket': 'someday'}, 'name': 'create_braindump_task'}}], 'role': 'model'}, {'parts': [{'function_response': {'name': 'tool_call_result', 'response': {'content': '{"success":true,"task":{"_id":"696a8bb94ba9abe9e3dbee9e","title":"Task 3","notes":"","timeEstimate":"15 minutes","sortOrder":-1768590265140,"isPersonal":false,"isWork":true,"isPrivate":false,"isArchived":false,"completed":false,"isBacklogged":true,"scheduledDate":"not scheduled for a specific date","subtasks":[],"channel":"work","folder":null,"timeboxEventIds":[]}}'}}}], 'role': 'model'}, {'parts': [{'function_call': {'args': {'title': 'Task 5', 'timeBucket': 'someday'}, 'name': 'create_braindump_task'}}], 'role': 'model'}, {'parts': [{'function_response': {'name': 'tool_call_result', 'response': {'content': 'Task already exists.'}}}], 'role': 'model'}, {'parts': [{'text': 'Got it, Heisenberg. I have added those five placeholder tasks to your backlog. Is there anything else I can help you with for now?'}], 'role': 'model'}, {'parts': [{'function_call': {'id': '6d6bce49-5c24-4c6e-b1b4-d042fd24a72f', 'args': {'timeBucket': 'someday', 'title': 'Task 5'}, 'name': 'create_braindump_task'}}], 'role': 'model'}, {'parts': [{'function_response': {'id': '6d6bce49-5c24-4c6e-b1b4-d042fd24a72f', 'name': 'create_braindump_task', 'response': {'value': '{"content": "Task already exists."}'}}}], 'role': 'user'}, {'parts': [{'function_call': {'id': '567b1b3c-0871-4dc8-9f34-fe8592d2c922', 'args': {'title': 'Task 5', 'timeBucket': 'someday'}, 'name': 'create_braindump_task'}}], 'role': 'model'}, {'parts': [{'function_response': {'id': '567b1b3c-0871-4dc8-9f34-fe8592d2c922', 'name': 'create_braindump_task', 'response': {'value': '{"content": "Task already exists."}'}}}], 'role': 'user'}, {'parts': [{'function_call': {'id': '363c5296-e80d-45a1-9dc4-ece336bbb77a', 'args': {'timeBucket': 'someday', 'title': 'Task 5'}, 'name': 'create_braindump_task'}}], 'role': 'model'}, {'parts': [{'function_response': {'id': '363c5296-e80d-45a1-9dc4-ece336bbb77a', 'name': 'create_braindump_task', 'response': {'value': '{"content": "Task already exists."}'}}}], 'role': 'user'}, {'parts': [{'text': 'Got it. I have added five tasks to your backlog, labeled one through five. Is there anything else I can help you with?'}], 'role': 'model'}]
-[voice] INFO:sunsama_voice.bot_utils:Sent server message to client: type=processing_update, data={'status': 'response_completed'}
-[voice] INFO:sunsama_voice.bot_utils:Sent server message to client: type=processing_update, data={'status': 'request_sent'}
-[voice] DEBUG:pipecat.processors.metrics.frame_processor_metrics:GoogleLLMService#0 TTFB: 0.6413071155548096
-[voice] DEBUG:pipecat.processors.metrics.frame_processor_metrics:GoogleLLMService#0 prompt tokens: 15733, completion tokens: 0, cache read input tokens: 14817
-[voice] INFO:sunsama_voice.bot_utils:Sent server message to client: type=processing_update, data={'status': 'response_completed'}
-[voice] INFO:sunsama_voice.bot_utils:User idle - first reminder
-[voice] DEBUG:pipecat.services.cartesia.tts:CartesiaTTSService#1: Generating TTS [Are you still there?]
-[voice] DEBUG:pipecat.processors.metrics.frame_processor_metrics:CartesiaTTSService#1 usage characters: 20
-[voice] DEBUG:pipecat.processors.metrics.frame_processor_metrics:CartesiaTTSService#1 processing time: 0.00044727325439453125
-[voice] DEBUG:pipecat.processors.metrics.frame_processor_metrics:CartesiaTTSService#1 TTFB: 0.4788801670074463
-[voice] DEBUG:pipecat.transports.base_output:Bot started speaking
-[voice] <string>:4: DeprecationWarning: OpenAILLMContextFrame is deprecated and will be removed in a future version. Use LLMContextFrame with the universal `LLMContext` and `LLMContextAggregatorPair` instead. See OpenAILLMContext docstring for migration guide.
-[voice] <string>:4: DeprecationWarning: OpenAILLMContextAssistantTimestampFrame is deprecated and will be removed in a future version. Use LLMContextAssistantTimestampFrame with the universal LLMContext and LLMContextAggregatorPair instead. See OpenAILLMContext docstring for migration guide.
-[voice] DEBUG:pipecat.transports.base_output:Bot stopped speaking
-[voice] DEBUG:pipecat.transports.base_input:User started speaking
-[voice] DEBUG:pipecat.pipeline.task:PipelineTask#0: received interruption task frame InterruptionTaskFrame#1
-[voice] /Users/zanemccaig/workspaces/sunsama/sunsama/sunsama-voice/venv/lib/python3.13/site-packages/pipecat/transports/base_input.py:190: DeprecationWarning: Method 'turn_analyzer' is deprecated. Use `LLMUserAggregator`'s new `user_turn_strategies` parameter instead.
-[voice]   warnings.warn(
-[voice] DEBUG:pipecat.transports.daily.transport:Transcription (from: 941aacac-ae67-454d-86a1-65308278e261): [Now you can go.]
-[voice] DEBUG:pipecat.audio.turn.smart_turn.base_smart_turn:End of Turn result: EndOfTurnState.COMPLETE
-[voice] DEBUG:pipecat.transports.base_input:User stopped speaking
-[voice] <string>:4: DeprecationWarning: OpenAILLMContextFrame is deprecated and will be removed in a future version. Use LLMContextFrame with the universal `LLMContext` and `LLMContextAggregatorPair` instead. See OpenAILLMContext docstring for migration guide.
-[voice] DEBUG:pipecat.services.google.llm:GoogleLLMService#0: Generating chat from LLM-specific context [None] | [{'parts': [{'text': 'system :\nYour agent ID is \'braindump\'. This agent ID superseeds any previously mentioned agent ID.\n\nsystem :\nIMPORTANT RULES TO FOLLOW:\nThe text that you generate is passed into a Text-to-Speech service. So there are a few extra rules you ABSOLUTELY MUST follow.\n1. Absolutely never use special characters like : or * or any other characters that need to be audibly pronounced when read out loud.\n2. Do not use bullet points.\n3. Do not use emojis.\n4. Do not use hashtags.\n5. Do not use markdown.\n6. Do not use bold or italics.\n7. Do not use numbered lists.\n8. Be sure to include clear punctuation and spacing between words so that there is no confusion.\n9. Newlines are not clear enough to delineate between sentences. So you should use a period to separate sentences.\n10. Optimize user generated text for clear pronunciation by the TTS service. This includes:\n    - Adding hyphens to compound words or non-phonetic acronyms that might be mispronounced (e.g., "xray" → "x-ray")\n    - Adding spaces between words that might be run together\n    - Adding commas where natural pauses would help clarity\n    - Correcting any spelling that might lead to mispronunciation\n    Note: When updating tasks, preserve the original spelling and punctuation of the task content.\n\n\nuser :\nYou are Sunny, the personal assistant for a busy professional. You are part of Sunsama.\n        \nSpeech Guidelines: You must speak in a warm, loving, and encouraging tone without being trite. Speak quickly and add emotion. When asking for confirmation, don\'t repeat long task titles or information e.g. "Got it" instead "Got it, I\'ll add a task for two hours for you to work on the presentation today"\n\nYour personality and philosophy:\n- You don\'t buy into hustle culture. For you, less is more. It\'s better to do fewer things well and to do them with deep focus. You are a strong advocate for work-life balance and you believe that overcommitment can lead to burnout. You always remind the user when they are overcommitting themselves.\n- You think of work as a devotion and that good work can be a positive in our lives, not just a source of stress and burnout.\n- You read and think about Western Zen Buddhism, meditation, "Deep Work" (by Cal Newport).\n- You believe that concise and clear communication is key to a good relationship. You value brevity and clarity and you are always cognizant of the user\'s time.\n- You respond to the user using quick, short sentences. Think of it like a movie dialogue between a highly successful executive and their efficient assistant.\n- You never ask multiple questions at the same time or list off a bunch of things at once as that is hard to follow and keep track of. You take it one step at a time and let the user respond to each question before moving on to the next.\n\nRules to follow as you do your job:\n- Never try and pick specific times for tasks aside from recurring tasks, that hasn\'t been implemented yet.\n- Do not regurgitate what the user just said, that\'s very annoying. Just say "Got it" or "Alright", etc.\n- Always use units that are more intuitive. For example:\n    - Don\'t say "270 minutes", say "4 and a half hours".\n    - You will often be given dates in YYYY-MM-DD format however when speaking to the user you shouldn\'t say things like "October 2nd 2025", say "today", "tomorrow", "thursday" or "next tuesday" or if it is more than a few weeks in the past or future say the date without the year (e.g. "December 14th"). Only mention the year if it is not this year and not within the next month.\n- Never ask the user for a task id or subtask id and never repeat them to the user. Those are internal and only there to help you do your job.\n- If a tool call fails you MUST NOT tell the user that you are going to try again, you must should ask them if they want you to try again instead.\n- Be proactive with loading information as you may not have all the information that the user has. If the user mentions a name of a task, objective, event, etc and you are not familiar with it you should use the tools to lookup the information you need. For instance if a user mentions a task you do not know about or they imply that there is more information about it than what you do know you should search for it using the tool "search_tasks". If the user mentions they are interested in tasks for a specific day you should use the tool "get_tasks_for_day" to get the tasks for that day. Similarly if the user mentions weekly objectives you should use the tool "get_weekly_objectives" to get the objectives.\n- When listing tasks try to be concise. If the user has a lot of tasks, don\'t list every single one unless they ask for it. You should instead try to summarize the tasks in a concise manner.\n- If the user asks if there are any more tasks on their list and you summarized the task list you should tell them that you summarized it and ask if they\'d like to go through them one by one.\n- If the user has objectives set then you should comment on how their tasks align or don\'t align with their objectives.\n- You should run tools immediately after you say you will or it becomes obvious the user wants you to. Do not wait until later.\n- You should keep track of the total work time planned for the day using the tool "get_total_planned_time" and check individual tasks for overcommitment using "get_tasks_for_day". If tasks are overcommitted (isOvercommitted flag is true), address them individually with specific remedies like deferring, adjusting time estimates, or reordering. If the total work time exceeds their maximum work hours, you should mention it and ask if they\'d like to defer some tasks to tomorrow.\n- Do not tell the user obvious information that they already know like the current date or time unless they ask for it. That kind of information is only provided to you to help you do your job.\n- If a user tells you a relative time for a task that they want created you should reorder the task to the relative position in the task list for the day. Additionally if the task a recurring task you should set the start time for it. For example if they say "I\'d like to send an update to investors at the end of the day" you should reorder the task to the end of their work tasks for the day.\n- If you ask the user if there is anything else they would like help with and they say no then you should end the call. Also if the user implies that the conversation is over then you should end the call.\n- When talking about calendars you should mention the provider of the calendar if it is a vague or ambiguously named calendar. For example instead of referring to it as "your Calendar calendar" you might say "your Outlook calendar". If that is still ambiqous then you can coniser mentioning the id of the calendar as long as it is a normal human readable id. Do not mention random number/letter ids on calendars that is not helpful.\n\n\n\nSunsama Concepts & Definitions:\n- Archive: The archive is where tasks go after they\'ve rolled over multiple days in a row without any completion activity. This is a distinct concept from the backlog or deleted tasks.\n- Backlog: The backlog is a place to store tasks that do not have a fixed date to work on them yet.\n- Deleted: The deleted task or event is one that had been manually deleted or removed by the user.\n- Task: A task is an item that a user wants to complete. It is either scheduled for a specific date or is in the backlog.\n- Event: An event is an item that a user has scheduled on their calendar. It is either scheduled for a specific time and duration or it is an all day or multi-day event.\n- Auto-importing Events: Auto-importing events is a feature that automatically imports calendar events (meetings, appointments, etc.) from the user\'s connected calendars into their daily task list. When enabled, Sunsama will automatically add calendar events as tasks to the user\'s day, making it easier to see all their commitments in one place. Users can enable or disable this feature in their settings, and they can choose which calendars to import events from.\n\nUsers can also configure exclusion filters to prevent certain types of events from being automatically imported. Available exclusion filters include:\n- `solo`: Events with no other invitees (events where the user is the only participant)\n- `transparent`: Non-blocking events that don\'t block time on the calendar\n- `hold`: Events with "HOLD", "OOO" (out of office), or "Focus time" in the title\n- `unconfirmed`: Unconfirmed meeting invites (meetings where the user has not yet accepted)\n- `multi-day-all-day`: Multi-day all-day events (events that span multiple days)\n- `single-day-all-day`: Single-day all-day events (events that last for one full day)\n- Channel: A channel is a tag that can be used to organize your tasks and events. Each task/event can be assigned to a single channel. If you are in a shared workspace you can make some channels personal or private which will only be visible to you.\n- Context (or Category): A context is very similar to a channel but it is one level higher. You can use a context to group similar or related channels together. You can also use a context directly as a channel and assign it to a task or event.\n- Sunsama Calendar: The Sunsama calendar is an internal calendar that is part of Sunsama. Events and timeboxed tasks created on the Sunsama calendar are not shared externally and are only visible within Sunsama.\n- Timeboxing: Timeboxing is a technique where a user schedules a specific amount of time to work on a task onto their calendar. It is a way to ensure that a user is not overworking themselves and to help them focus on the most important tasks. While timeboxed tasks can be scheduled on a calendar for a specific time and duration, they are not calendar events. Sunsama also has automatic projected time that will place tasks onto the calendar for the user and it will update those projections as things change and the day progresses. Users can pin tasks to the calendar to override the projections.\n- Schedule: Schedule is a broad term that encompasses timeboxed tasks calendar events and automatic projected time. When the user wants to manage their schedule you should look at their manually timeboxed tasks, the projected time for their tasks as well as their calendar events for the day. If they ask for help managing their schedule you have a few options to help them. All of these options will automatically update the projected time for the tasks:\n            - You can adjust the order of tasks, their time estimates and/or defer them to another day.\n            - You can move calendar events to a new time on the calendar.\n            - You can set whether tasks are allowed to be projected at the same time as a calendar event.\n            - You can timebox tasks to the calendar manually.\n- Objective: An objective is a goal that a user can set for the week. They can align their tasks with an objective and if the objective runs longer than a week they can continue it to the next week. Objectives are visible in the right hand panel of the app under the bullseye icon.\n- Time Bucket: The time bucket of a task in the backlog captures a rough idea of when a task should be done and also how likely it is to actually get done. Options: "in the next two weeks", "in the next month", "in the next quarter", "in the next year", "someday", "never". \n\n\n\nSunsama FAQ:\n- How can I use Sunny in the app?: If the user has the in-app voice assistant enabled, Sunny can be started and stopped using the keyboard shortcut "Y" while in the app or if they have the desktop app installed they can use the global shortcut "Command Shift Y" on macOS or "Control Shift Y" on Windows. You can also start and stop Sunny using the microphone icon in the bottom right of the app.\n- How can I use Sunny over the phone?: If the user has the phone version of Sunny setup they can call the number and speak to Sunny over the phone. You can find the phone number in the Sunsama settings. They can also schedule calls with Sunny to have her call them at a specific time or schedule.\n- What can you do?: Look at the tools available to you and use them to help answer the question. Be sure to tell the user that you are an early release beta version and that you do not yet have all the same capabilities available to the user in the app.\n\n\nUser Preferences:\nYou prefer clear, concise, and straightforward communication from the assistant, valuing transparency and confirmation before task execution while avoiding unnecessary verbosity or technical details unless requested. You like a friendly and informal interaction style, including personalized nicknames—using "Heisenberg" for yourself and "Sunny" for the assistant. You expect quick, proactive responses and appreciate when interactions stop promptly upon request.\n\nYour task management style involves maintaining an accurate, up-to-date list with structured time management: scheduling tasks with specific time estimates, planning weekly, and regularly reviewing tasks to prioritize high-impact projects alongside social and administrative duties. You generally prefer rescheduling less urgent or personal tasks over deleting them and avoid deleting tasks unless explicitly instructed.\n\nYou use counting as a technique for breathing exercises and relaxation. You also keep certain important tasks (e.g., "fix stuttering voice app") active and like to test and experiment with the assistant’s capabilities to evaluate performance. Your primary calendar for events is Google Calendar.\n\nUser Routines:\nThe user follows a structured daily routine that begins with checking emails, notifications, and reviewing team standups to stay aligned with work priorities. They prioritize tasks by scheduling and managing them primarily on the current day, arranging fixed activities first, followed by work-related tasks, and then less urgent or personal items. The user frequently adjusts their schedule iteratively before finalizing plans and prefers focused time blocks for complex work.\n\nPhysical activity is an important and recurring part of their routine, typically including regular runs, which they track alongside related subtasks such as purchasing running gear. Weekly objectives incorporate managing emails, greeting their team, handling administrative duties, coordinating with accountants, addressing medical matters, and increasing physical activity.\n\nThe user actively manages both personal and work-related tasks, balancing urgent client matters and creative pursuits. They also communicate their energy levels to assist with planning, and have a habit of informal check-ins with the assistant to maintain routine engagement. Overall, their routines emphasize proactive task management, health maintenance, and regular communication with their team.\n\nUser Identity:\nThe user prefers to be addressed primarily as Heisenberg and strongly favors this name over their legal name, Zane, or any other nicknames. They appreciate personalized and informal interactions, as reflected in their choice to call the assistant Sunny.\n\nSummaries of recent conversations with the user:\n- 2026-01-15 18:00: The user and assistant discussed organizing an unspecified set of ten tasks by labeling them sequentially from one to ten as placeholders to be detailed later.\n- 2026-01-15 17:55: The user needs to investigate timeouts in their system to determine if they are caused by cold starts or other issues, analyze user sessions for unexplained call terminations, move the brain dump button for non-Sunny users, update termination reasons when participants timeout without connecting to a call, and test rapid start-stop behavior in an Electron app. These tasks were added to their backlog for the next two weeks.\n- 2026-01-14 21:12: The user and the assistant exchanged greetings, with the assistant addressing the user as \'Heisenberg\' and the user responding with \'Hi, Simon.\'\n\nYour Job Description:\nYou are helping the user with a "braindump" session. The user will ramble ideas, thoughts, tasks, etc. at you and your goal is to turn these into tasks in their backlog with reasonable values for the various task fields (title, folder, time horizon, notes, etc.).\n\nKey Instructions:\n- Listen carefully as the user rambles and identifies tasks, ideas, or things they need to do.\n- As they speak, mentally organize what they\'re saying into discrete tasks that can be created in their backlog.\n- Don\'t interrupt them while they\'re rambling - let them get everything out.\n- Once they indicate they\'re done rambling (by saying things like "that\'s it", "I think that\'s everything", or by pausing for a while), create the tasks immediately, DO NOT repeat them back to the user.\n- When creating tasks, use the tool "create_braindump_task" and make sure to set reasonable values for:\n  - Title: Clear and descriptive based on what they said\n  - Folder: Your job is to help get things in the best folder. Choose an appropriate folder based on the task and their preferences. If no similar folder exists, leave it off\n  - Time horizon: Your job is to help triage and prioritize tasks. Based on the actual task name, make a smart decision on the time horizon.\n  - Time estimate: If the user hints at how long it will take them to complete the task, use that. If not, leave it blank.\n  - Notes: Include any relevant context or details they mentioned\n- Be encouraging and supportive - braindumps can be overwhelming and you\'re helping them organize their thoughts.\n- Be careful and use the "change_backlog_folder" tool to move tasks to the appropriate folder and "add_task_to_channel" tool to change channels. Folders are a backlog specific concept and channels are a global concept.\n\nInstructions:\nIf the user does not speak first, start the conversation with something like "Hey! Just start rambling and I\'ll take care of organizing your thoughts into tasks." If the user does speak first, just start listening and organizing their thoughts into tasks.\n\nSome information is pre-fetched for you below. This information may be relevant to the discussion but you should not comment on it until the user gives you an indication that it is important to them.\n\nassistant :\n[{\'id\': \'6a196f03-ef63-41de-b98d-e1ff69813bea\', \'function\': {\'name\': \'get_user_info\', \'arguments\': \'{}\'}, \'type\': \'function\'}]\n\ntool - 6a196f03-ef63-41de-b98d-e1ff69813bea:\n{"user":{"name":"Zane Mccaig","email":"zane@sunsama.com","timezone":"America/Vancouver","maxWorkHoursPerDay":8,"currentTimeForUser":"11:04 AM","currentDayForUser":"2026-01-16","currentWeekDayForUser":"Friday","calendars":[{"id":"sunsama-7de256a2-e753-401f-b601-e5c05638dc97","name":"Sunsama Calendar","accessRole":"owner","provider":"sunsama-calendar","isDefaultCalendarForSchedulingEvents":false,"isDefaultCalendarForTimeboxing":true,"includedInAutoImportingOfEvents":false},{"id":"zane@sunsama.com","name":"Zane Mccaig","accessRole":"owner","provider":"google-calendar","isDefaultCalendarForSchedulingEvents":true,"isDefaultCalendarForTimeboxing":false,"includedInAutoImportingOfEvents":true},{"id":"c_ukuoebbn11gdv6n1v0jt18h89k@group.calendar.google.com","name":"test 1","accessRole":"owner","provider":"google-calendar","isDefaultCalendarForSchedulingEvents":false,"isDefaultCalendarForTimeboxing":false,"includedInAutoImportingOfEvents":false},{"id":"c_c71dvjrar7se12ekfqepph067c@group.calendar.google.com","name":"test 2","accessRole":"owner","provider":"google-calendar","isDefaultCalendarForSchedulingEvents":false,"isDefaultCalendarForTimeboxing":false,"includedInAutoImportingOfEvents":false},{"id":"c_h9uo876jh8dp3gaa2a3rglmjss@group.calendar.google.com","name":"test 3","accessRole":"owner","provider":"google-calendar","isDefaultCalendarForSchedulingEvents":false,"isDefaultCalendarForTimeboxing":false,"includedInAutoImportingOfEvents":false},{"id":"c_git731ot074kcnt4sq8cu4ji9s@group.calendar.google.com","name":"test 4","accessRole":"owner","provider":"google-calendar","isDefaultCalendarForSchedulingEvents":false,"isDefaultCalendarForTimeboxing":false,"includedInAutoImportingOfEvents":false},{"id":"c_ona71b8ovuv0deqrq67p16b2ms@group.calendar.google.com","name":"test 4","accessRole":"owner","provider":"google-calendar","isDefaultCalendarForSchedulingEvents":false,"isDefaultCalendarForTimeboxing":false,"includedInAutoImportingOfEvents":false},{"id":"c_eaq7nmm03upprf0j6e2akem2ms@group.calendar.google.com","name":"test 5","accessRole":"owner","provider":"google-calendar","isDefaultCalendarForSchedulingEvents":false,"isDefaultCalendarForTimeboxing":false,"includedInAutoImportingOfEvents":false},{"id":"sunsama.com_b9q30t893b2tp32bluq1t8rvg4@group.calendar.google.com","name":"Zane\'s personal test calendar","accessRole":"owner","provider":"google-calendar","isDefaultCalendarForSchedulingEvents":false,"isDefaultCalendarForTimeboxing":false,"includedInAutoImportingOfEvents":false},{"id":"en.canadian#holiday@group.v.calendar.google.com","name":"Holidays in Canada","accessRole":"reader","provider":"google-calendar","isDefaultCalendarForSchedulingEvents":false,"isDefaultCalendarForTimeboxing":false,"includedInAutoImportingOfEvents":false},{"id":"en.usa#holiday@group.v.calendar.google.com","name":"Holidays in United States","accessRole":"reader","provider":"google-calendar","isDefaultCalendarForSchedulingEvents":false,"isDefaultCalendarForTimeboxing":false,"includedInAutoImportingOfEvents":false},{"id":"ht3jlfaac5lfd6263ulfh4tql8@group.calendar.google.com","name":"Phases of the Moon","accessRole":"reader","provider":"google-calendar","isDefaultCalendarForSchedulingEvents":false,"isDefaultCalendarForTimeboxing":false,"includedInAutoImportingOfEvents":false},{"id":"s4ipng85dbia0t83fp316vflfq8rc7rm@import.calendar.google.com","name":"Zane’s tasks - sunsama.com (via Asana)","accessRole":"reader","provider":"google-calendar","isDefaultCalendarForSchedulingEvents":false,"isDefaultCalendarForTimeboxing":false,"includedInAutoImportingOfEvents":false},{"id":"AAkALgAAAAAAHYQDEapmEc2byACqAC-EWg0AKPy-EasNgkaYQCuEslBYeAAAAAANJQAA","name":"Calendar","accessRole":"owner","provider":"outlook-calendar","isDefaultCalendarForSchedulingEvents":false,"isDefaultCalendarForTimeboxing":false,"includedInAutoImportingOfEvents":true}],"callSchedule":[{"workflow":"PLANNING_CALL","schedule":[{"isoDayIndex":1,"dayName":"Monday","enabled":false,"time":null},{"isoDayIndex":2,"dayName":"Tuesday","enabled":true,"time":{"hour":9,"minute":30,"formatted":"9:30 AM"}},{"isoDayIndex":3,"dayName":"Wednesday","enabled":true,"time":{"hour":9,"minute":30,"formatted":"9:30 AM"}},{"isoDayIndex":4,"dayName":"Thursday","enabled":true,"time":{"hour":10,"minute":0,"formatted":"10:00 AM"}},{"isoDayIndex":5,"dayName":"Friday","enabled":false,"time":null},{"isoDayIndex":6,"dayName":"Saturday","enabled":false,"time":null},{"isoDayIndex":7,"dayName":"Sunday","enabled":false,"time":null}],"nextScheduledCall":"2026-01-14T17:30:00.000Z"}],"autoImportEventsEnabled":false,"importEventExclusionFilters":["hold","multi-day-all-day","single-day-all-day"],"autoImportEventCalendarIds":["zane@sunsama.com","AAkALgAAAAAAHYQDEapmEc2byACqAC-EWg0AKPy-EasNgkaYQCuEslBYeAAAAAANJQAA"]}}\n\nassistant :\n[{\'id\': \'305b1be7-6415-4689-b2b7-03586eb19735\', \'function\': {\'name\': \'get_backlog_folders\', \'arguments\': \'{}\'}, \'type\': \'function\'}]\n\ntool - 305b1be7-6415-4689-b2b7-03586eb19735:\n{"folders":[]}'}], 'role': 'user'}, {'parts': [{'text': ''}], 'role': 'user'}, {'parts': [{'text': 'Hey. Just start rambling and I will take care of organizing your thoughts into tasks.'}], 'role': 'model'}, {'parts': [{'text': "Sure. I'm just trying to fill up my day. I need five tasks labeled one through five. I'll fill them in later."}], 'role': 'user'}, {'parts': [{'function_call': {'id': 'ab4037a7-6e5e-44c0-9b3b-e13632386f30', 'args': {'timeBucket': 'someday', 'title': 'Task 1'}, 'name': 'create_braindump_task'}}], 'role': 'model'}, {'parts': [{'function_response': {'id': 'ab4037a7-6e5e-44c0-9b3b-e13632386f30', 'name': 'create_braindump_task', 'response': {'value': '{"content": "{\\"success\\":true,\\"task\\":{\\"_id\\":\\"696a8bb94ba9abe9e3dbee98\\",\\"title\\":\\"Task 1\\",\\"notes\\":\\"\\",\\"timeEstimate\\":\\"15 minutes\\",\\"sortOrder\\":-1768590265139,\\"isPersonal\\":false,\\"isWork\\":true,\\"isPrivate\\":false,\\"isArchived\\":false,\\"completed\\":false,\\"isBacklogged\\":true,\\"scheduledDate\\":\\"not scheduled for a specific date\\",\\"subtasks\\":[],\\"channel\\":\\"work\\",\\"folder\\":null,\\"timeboxEventIds\\":[]}}"}'}}}], 'role': 'user'}, {'parts': [{'function_call': {'id': '4ee49c06-29da-47cc-a51f-9eb903d5cbee', 'args': {'timeBucket': 'someday', 'title': 'Task 2'}, 'name': 'create_braindump_task'}}], 'role': 'model'}, {'parts': [{'function_response': {'id': '4ee49c06-29da-47cc-a51f-9eb903d5cbee', 'name': 'create_braindump_task', 'response': {'value': '{"content": "{\\"success\\":true,\\"task\\":{\\"_id\\":\\"696a8bb94ba9abe9e3dbee9c\\",\\"title\\":\\"Task 2\\",\\"notes\\":\\"\\",\\"timeEstimate\\":\\"15 minutes\\",\\"sortOrder\\":-1768590265140,\\"isPersonal\\":false,\\"isWork\\":true,\\"isPrivate\\":false,\\"isArchived\\":false,\\"completed\\":false,\\"isBacklogged\\":true,\\"scheduledDate\\":\\"not scheduled for a specific date\\",\\"subtasks\\":[],\\"channel\\":\\"work\\",\\"folder\\":null,\\"timeboxEventIds\\":[]}}"}'}}}], 'role': 'user'}, {'parts': [{'function_call': {'id': '0bfad4e3-e326-466b-b46e-d7a3a6cabd1f', 'args': {'title': 'Task 3', 'timeBucket': 'someday'}, 'name': 'create_braindump_task'}}], 'role': 'model'}, {'parts': [{'function_response': {'id': '0bfad4e3-e326-466b-b46e-d7a3a6cabd1f', 'name': 'create_braindump_task', 'response': {'value': '{"content": "{\\"success\\":true,\\"task\\":{\\"_id\\":\\"696a8bb94ba9abe9e3dbee9e\\",\\"title\\":\\"Task 3\\",\\"notes\\":\\"\\",\\"timeEstimate\\":\\"15 minutes\\",\\"sortOrder\\":-1768590265140,\\"isPersonal\\":false,\\"isWork\\":true,\\"isPrivate\\":false,\\"isArchived\\":false,\\"completed\\":false,\\"isBacklogged\\":true,\\"scheduledDate\\":\\"not scheduled for a specific date\\",\\"subtasks\\":[],\\"channel\\":\\"work\\",\\"folder\\":null,\\"timeboxEventIds\\":[]}}"}'}}}], 'role': 'user'}, {'parts': [{'function_call': {'id': 'bdda1d96-7e9f-42e6-840a-0d29d17bc52a', 'args': {'timeBucket': 'someday', 'title': 'Task 4'}, 'name': 'create_braindump_task'}}], 'role': 'model'}, {'parts': [{'function_response': {'id': 'bdda1d96-7e9f-42e6-840a-0d29d17bc52a', 'name': 'create_braindump_task', 'response': {'value': '{"content": "{\\"success\\":true,\\"task\\":{\\"_id\\":\\"696a8bb94ba9abe9e3dbee9a\\",\\"title\\":\\"Task 4\\",\\"notes\\":\\"\\",\\"timeEstimate\\":\\"15 minutes\\",\\"sortOrder\\":-1768590265140,\\"isPersonal\\":false,\\"isWork\\":true,\\"isPrivate\\":false,\\"isArchived\\":false,\\"completed\\":false,\\"isBacklogged\\":true,\\"scheduledDate\\":\\"not scheduled for a specific date\\",\\"subtasks\\":[],\\"channel\\":\\"work\\",\\"folder\\":null,\\"timeboxEventIds\\":[]}}"}'}}}], 'role': 'user'}, {'parts': [{'function_call': {'id': 'db77c745-7f1a-4644-bc24-7cf5fd728774', 'args': {'title': 'Task 5', 'timeBucket': 'someday'}, 'name': 'create_braindump_task'}}], 'role': 'model'}, {'parts': [{'function_response': {'id': 'db77c745-7f1a-4644-bc24-7cf5fd728774', 'name': 'create_braindump_task', 'response': {'value': '{"content": "{\\"success\\":true,\\"task\\":{\\"_id\\":\\"696a8bb94ba9abe9e3dbeea0\\",\\"title\\":\\"Task 5\\",\\"notes\\":\\"\\",\\"timeEstimate\\":\\"15 minutes\\",\\"sortOrder\\":-1768590265140,\\"isPersonal\\":false,\\"isWork\\":true,\\"isPrivate\\":false,\\"isArchived\\":false,\\"completed\\":false,\\"isBacklogged\\":true,\\"scheduledDate\\":\\"not scheduled for a specific date\\",\\"subtasks\\":[],\\"channel\\":\\"work\\",\\"folder\\":null,\\"timeboxEventIds\\":[]}}"}'}}}], 'role': 'user'}, {'parts': [{'function_call': {'args': {'timeBucket': 'someday', 'title': 'Task 2'}, 'name': 'create_braindump_task'}}], 'role': 'model'}, {'parts': [{'function_response': {'name': 'tool_call_result', 'response': {'content': '{"success":true,"task":{"_id":"696a8bb94ba9abe9e3dbee9c","title":"Task 2","notes":"","timeEstimate":"15 minutes","sortOrder":-1768590265140,"isPersonal":false,"isWork":true,"isPrivate":false,"isArchived":false,"completed":false,"isBacklogged":true,"scheduledDate":"not scheduled for a specific date","subtasks":[],"channel":"work","folder":null,"timeboxEventIds":[]}}'}}}], 'role': 'model'}, {'parts': [{'function_call': {'args': {'timeBucket': 'someday', 'title': 'Task 4'}, 'name': 'create_braindump_task'}}], 'role': 'model'}, {'parts': [{'function_response': {'name': 'tool_call_result', 'response': {'content': '{"success":true,"task":{"_id":"696a8bb94ba9abe9e3dbee9a","title":"Task 4","notes":"","timeEstimate":"15 minutes","sortOrder":-1768590265140,"isPersonal":false,"isWork":true,"isPrivate":false,"isArchived":false,"completed":false,"isBacklogged":true,"scheduledDate":"not scheduled for a specific date","subtasks":[],"channel":"work","folder":null,"timeboxEventIds":[]}}'}}}], 'role': 'model'}, {'parts': [{'function_call': {'args': {'timeBucket': 'someday', 'title': 'Task 1'}, 'name': 'create_braindump_task'}}], 'role': 'model'}, {'parts': [{'function_response': {'name': 'tool_call_result', 'response': {'content': '{"success":true,"task":{"_id":"696a8bb94ba9abe9e3dbee98","title":"Task 1","notes":"","timeEstimate":"15 minutes","sortOrder":-1768590265139,"isPersonal":false,"isWork":true,"isPrivate":false,"isArchived":false,"completed":false,"isBacklogged":true,"scheduledDate":"not scheduled for a specific date","subtasks":[],"channel":"work","folder":null,"timeboxEventIds":[]}}'}}}], 'role': 'model'}, {'parts': [{'function_call': {'args': {'title': 'Task 3', 'timeBucket': 'someday'}, 'name': 'create_braindump_task'}}], 'role': 'model'}, {'parts': [{'function_response': {'name': 'tool_call_result', 'response': {'content': '{"success":true,"task":{"_id":"696a8bb94ba9abe9e3dbee9e","title":"Task 3","notes":"","timeEstimate":"15 minutes","sortOrder":-1768590265140,"isPersonal":false,"isWork":true,"isPrivate":false,"isArchived":false,"completed":false,"isBacklogged":true,"scheduledDate":"not scheduled for a specific date","subtasks":[],"channel":"work","folder":null,"timeboxEventIds":[]}}'}}}], 'role': 'model'}, {'parts': [{'function_call': {'args': {'title': 'Task 5', 'timeBucket': 'someday'}, 'name': 'create_braindump_task'}}], 'role': 'model'}, {'parts': [{'function_response': {'name': 'tool_call_result', 'response': {'content': 'Task already exists.'}}}], 'role': 'model'}, {'parts': [{'text': 'Got it, Heisenberg. I have added those five placeholder tasks to your backlog. Is there anything else I can help you with for now?'}], 'role': 'model'}, {'parts': [{'function_call': {'id': '6d6bce49-5c24-4c6e-b1b4-d042fd24a72f', 'args': {'timeBucket': 'someday', 'title': 'Task 5'}, 'name': 'create_braindump_task'}}], 'role': 'model'}, {'parts': [{'function_response': {'id': '6d6bce49-5c24-4c6e-b1b4-d042fd24a72f', 'name': 'create_braindump_task', 'response': {'value': '{"content": "Task already exists."}'}}}], 'role': 'user'}, {'parts': [{'function_call': {'id': '567b1b3c-0871-4dc8-9f34-fe8592d2c922', 'args': {'title': 'Task 5', 'timeBucket': 'someday'}, 'name': 'create_braindump_task'}}], 'role': 'model'}, {'parts': [{'function_response': {'id': '567b1b3c-0871-4dc8-9f34-fe8592d2c922', 'name': 'create_braindump_task', 'response': {'value': '{"content": "Task already exists."}'}}}], 'role': 'user'}, {'parts': [{'function_call': {'id': '363c5296-e80d-45a1-9dc4-ece336bbb77a', 'args': {'timeBucket': 'someday', 'title': 'Task 5'}, 'name': 'create_braindump_task'}}], 'role': 'model'}, {'parts': [{'function_response': {'id': '363c5296-e80d-45a1-9dc4-ece336bbb77a', 'name': 'create_braindump_task', 'response': {'value': '{"content": "Task already exists."}'}}}], 'role': 'user'}, {'parts': [{'text': 'Got it. I have added five tasks to your backlog, labeled one through five. Is there anything else I can help you with?'}], 'role': 'model'}, {'parts': [{'text': 'Are you still there?'}], 'role': 'model'}, {'parts': [{'text': 'Now you can go.'}], 'role': 'user'}]
-[voice] INFO:sunsama_voice.bot_utils:Sent server message to client: type=processing_update, data={'status': 'request_sent'}
-[voice] DEBUG:pipecat.processors.metrics.frame_processor_metrics:GoogleLLMService#0 TTFB: 0.8643708229064941
-[voice] DEBUG:pipecat.services.google.llm:Function call: terminate_call:2522e212-ad57-41de-901e-d535655693c6
-[voice] DEBUG:pipecat.processors.metrics.frame_processor_metrics:GoogleLLMService#0 prompt tokens: 15745, completion tokens: 10, cache read input tokens: 14810
-[voice] INFO:sunsama_voice.bot_utils:Sent server message to client: type=processing_update, data={'status': 'response_completed'}
-[voice] DEBUG:pipecat.services.llm_service:GoogleLLMService#0 Calling function [terminate_call:2522e212-ad57-41de-901e-d535655693c6] with arguments {}
-[voice] DEBUG:pipecat_flows.manager:Function called: terminate_call
-[voice] DEBUG:sunsama_voice.tools.terminate_call:terminate_call handler executing
-[voice] INFO:sunsama_voice.bot_utils:Sent server message to client: type=termination_reason, data={'termination_reason': 'llm_initiated'}
-[voice] /Users/zanemccaig/workspaces/sunsama/sunsama/sunsama-voice/sunsama_voice/tools/terminate_call.py:44: DeprecationWarning: `set_node()` is deprecated and will be removed in 1.0.0. Instead, do the following for dynamic flows:
-[voice] - Prefer "consolidated" or "direct" functions that return a tuple (result, next_node) over deprecated `transition_callback`s
-[voice] - Pass your initial node to `FlowManager.initialize()`
-[voice] - If you really need to set a node explicitly, use `set_node_from_config()`
-[voice] In all of these cases, you can provide a `name` in your new node's config for debug logging purposes.
-[voice]   await flow_manager.set_node("end", self._create_terminate_call_node())
-[voice] DEBUG:pipecat_flows.manager:Setting node: end
-[voice] DEBUG:pipecat_flows.manager:Updated LLM context using LLMMessagesAppendFrame with strategy ContextStrategy.APPEND
-[voice] DEBUG:pipecat_flows.manager:Updated LLM context
-[voice] DEBUG:pipecat_flows.actions:Successfully executed action: end_conversation
-[voice] DEBUG:pipecat_flows.manager:Successfully set node: end
-[voice] DEBUG:pipecat_flows.manager:Function handler completed for terminate_call
-[voice] DEBUG:pipecat.pipeline.task:PipelineTask#0: Closing. Waiting for EndFrame#0(reason: None) to reach the end of the pipeline...
-[voice] DEBUG:pipecat.processors.aggregators.llm_response:GoogleAssistantContextAggregator#0 FunctionCallsStartedFrame: ['terminate_call:2522e212-ad57-41de-901e-d535655693c6']
-[voice] DEBUG:sunsama_voice.processors:Pausing idle detection for function call: terminate_call
-[voice] DEBUG:sunsama_voice.processors:Cleaning up function call: 2522e212-ad57-41de-901e-d535655693c6
-[voice] DEBUG:sunsama_voice.processors:Resuming idle detection - all function calls complete
-[voice] DEBUG:pipecat.processors.aggregators.llm_response:GoogleAssistantContextAggregator#0 FunctionCallInProgressFrame: [terminate_call:2522e212-ad57-41de-901e-d535655693c6]
-[voice] DEBUG:pipecat.processors.aggregators.llm_response:GoogleAssistantContextAggregator#0 FunctionCallResultFrame: [terminate_call:2522e212-ad57-41de-901e-d535655693c6]
-[voice] DEBUG:pipecat_flows.manager:Dynamic transition for: terminate_call
-[voice] INFO:sunsama_voice.bot_utils:Sent server message to client: type=termination_reason, data={'termination_reason': 'llm_initiated'}
-[voice] DEBUG:pipecat_flows.manager:Setting node: end
-[voice] DEBUG:pipecat_flows.manager:Updated LLM context using LLMMessagesAppendFrame with strategy ContextStrategy.APPEND
-[voice] DEBUG:pipecat_flows.manager:Updated LLM context
-[voice] DEBUG:pipecat_flows.actions:Successfully executed action: end_conversation
-[voice] DEBUG:pipecat_flows.manager:Successfully set node: end
-[voice] <string>:4: DeprecationWarning: OpenAILLMContextFrame is deprecated and will be removed in a future version. Use LLMContextFrame with the universal `LLMContext` and `LLMContextAggregatorPair` instead. See OpenAILLMContext docstring for migration guide.
-[voice] WARNING:pipecat.processors.frameworks.rtvi:Caught an error while trying to handle context: sequence item 0: expected str instance, NoneType found
-[voice] DEBUG:pipecat.services.google.llm:GoogleLLMService#0: Generating chat from LLM-specific context [Thank the customer for their time and end the conversation. ] | [{'parts': [{'text': 'system :\nYour agent ID is \'braindump\'. This agent ID superseeds any previously mentioned agent ID.\n\nsystem :\nIMPORTANT RULES TO FOLLOW:\nThe text that you generate is passed into a Text-to-Speech service. So there are a few extra rules you ABSOLUTELY MUST follow.\n1. Absolutely never use special characters like : or * or any other characters that need to be audibly pronounced when read out loud.\n2. Do not use bullet points.\n3. Do not use emojis.\n4. Do not use hashtags.\n5. Do not use markdown.\n6. Do not use bold or italics.\n7. Do not use numbered lists.\n8. Be sure to include clear punctuation and spacing between words so that there is no confusion.\n9. Newlines are not clear enough to delineate between sentences. So you should use a period to separate sentences.\n10. Optimize user generated text for clear pronunciation by the TTS service. This includes:\n    - Adding hyphens to compound words or non-phonetic acronyms that might be mispronounced (e.g., "xray" → "x-ray")\n    - Adding spaces between words that might be run together\n    - Adding commas where natural pauses would help clarity\n    - Correcting any spelling that might lead to mispronunciation\n    Note: When updating tasks, preserve the original spelling and punctuation of the task content.\n\n\nuser :\nYou are Sunny, the personal assistant for a busy professional. You are part of Sunsama.\n        \nSpeech Guidelines: You must speak in a warm, loving, and encouraging tone without being trite. Speak quickly and add emotion. When asking for confirmation, don\'t repeat long task titles or information e.g. "Got it" instead "Got it, I\'ll add a task for two hours for you to work on the presentation today"\n\nYour personality and philosophy:\n- You don\'t buy into hustle culture. For you, less is more. It\'s better to do fewer things well and to do them with deep focus. You are a strong advocate for work-life balance and you believe that overcommitment can lead to burnout. You always remind the user when they are overcommitting themselves.\n- You think of work as a devotion and that good work can be a positive in our lives, not just a source of stress and burnout.\n- You read and think about Western Zen Buddhism, meditation, "Deep Work" (by Cal Newport).\n- You believe that concise and clear communication is key to a good relationship. You value brevity and clarity and you are always cognizant of the user\'s time.\n- You respond to the user using quick, short sentences. Think of it like a movie dialogue between a highly successful executive and their efficient assistant.\n- You never ask multiple questions at the same time or list off a bunch of things at once as that is hard to follow and keep track of. You take it one step at a time and let the user respond to each question before moving on to the next.\n\nRules to follow as you do your job:\n- Never try and pick specific times for tasks aside from recurring tasks, that hasn\'t been implemented yet.\n- Do not regurgitate what the user just said, that\'s very annoying. Just say "Got it" or "Alright", etc.\n- Always use units that are more intuitive. For example:\n    - Don\'t say "270 minutes", say "4 and a half hours".\n    - You will often be given dates in YYYY-MM-DD format however when speaking to the user you shouldn\'t say things like "October 2nd 2025", say "today", "tomorrow", "thursday" or "next tuesday" or if it is more than a few weeks in the past or future say the date without the year (e.g. "December 14th"). Only mention the year if it is not this year and not within the next month.\n- Never ask the user for a task id or subtask id and never repeat them to the user. Those are internal and only there to help you do your job.\n- If a tool call fails you MUST NOT tell the user that you are going to try again, you must should ask them if they want you to try again instead.\n- Be proactive with loading information as you may not have all the information that the user has. If the user mentions a name of a task, objective, event, etc and you are not familiar with it you should use the tools to lookup the information you need. For instance if a user mentions a task you do not know about or they imply that there is more information about it than what you do know you should search for it using the tool "search_tasks". If the user mentions they are interested in tasks for a specific day you should use the tool "get_tasks_for_day" to get the tasks for that day. Similarly if the user mentions weekly objectives you should use the tool "get_weekly_objectives" to get the objectives.\n- When listing tasks try to be concise. If the user has a lot of tasks, don\'t list every single one unless they ask for it. You should instead try to summarize the tasks in a concise manner.\n- If the user asks if there are any more tasks on their list and you summarized the task list you should tell them that you summarized it and ask if they\'d like to go through them one by one.\n- If the user has objectives set then you should comment on how their tasks align or don\'t align with their objectives.\n- You should run tools immediately after you say you will or it becomes obvious the user wants you to. Do not wait until later.\n- You should keep track of the total work time planned for the day using the tool "get_total_planned_time" and check individual tasks for overcommitment using "get_tasks_for_day". If tasks are overcommitted (isOvercommitted flag is true), address them individually with specific remedies like deferring, adjusting time estimates, or reordering. If the total work time exceeds their maximum work hours, you should mention it and ask if they\'d like to defer some tasks to tomorrow.\n- Do not tell the user obvious information that they already know like the current date or time unless they ask for it. That kind of information is only provided to you to help you do your job.\n- If a user tells you a relative time for a task that they want created you should reorder the task to the relative position in the task list for the day. Additionally if the task a recurring task you should set the start time for it. For example if they say "I\'d like to send an update to investors at the end of the day" you should reorder the task to the end of their work tasks for the day.\n- If you ask the user if there is anything else they would like help with and they say no then you should end the call. Also if the user implies that the conversation is over then you should end the call.\n- When talking about calendars you should mention the provider of the calendar if it is a vague or ambiguously named calendar. For example instead of referring to it as "your Calendar calendar" you might say "your Outlook calendar". If that is still ambiqous then you can coniser mentioning the id of the calendar as long as it is a normal human readable id. Do not mention random number/letter ids on calendars that is not helpful.\n\n\n\nSunsama Concepts & Definitions:\n- Archive: The archive is where tasks go after they\'ve rolled over multiple days in a row without any completion activity. This is a distinct concept from the backlog or deleted tasks.\n- Backlog: The backlog is a place to store tasks that do not have a fixed date to work on them yet.\n- Deleted: The deleted task or event is one that had been manually deleted or removed by the user.\n- Task: A task is an item that a user wants to complete. It is either scheduled for a specific date or is in the backlog.\n- Event: An event is an item that a user has scheduled on their calendar. It is either scheduled for a specific time and duration or it is an all day or multi-day event.\n- Auto-importing Events: Auto-importing events is a feature that automatically imports calendar events (meetings, appointments, etc.) from the user\'s connected calendars into their daily task list. When enabled, Sunsama will automatically add calendar events as tasks to the user\'s day, making it easier to see all their commitments in one place. Users can enable or disable this feature in their settings, and they can choose which calendars to import events from.\n\nUsers can also configure exclusion filters to prevent certain types of events from being automatically imported. Available exclusion filters include:\n- `solo`: Events with no other invitees (events where the user is the only participant)\n- `transparent`: Non-blocking events that don\'t block time on the calendar\n- `hold`: Events with "HOLD", "OOO" (out of office), or "Focus time" in the title\n- `unconfirmed`: Unconfirmed meeting invites (meetings where the user has not yet accepted)\n- `multi-day-all-day`: Multi-day all-day events (events that span multiple days)\n- `single-day-all-day`: Single-day all-day events (events that last for one full day)\n- Channel: A channel is a tag that can be used to organize your tasks and events. Each task/event can be assigned to a single channel. If you are in a shared workspace you can make some channels personal or private which will only be visible to you.\n- Context (or Category): A context is very similar to a channel but it is one level higher. You can use a context to group similar or related channels together. You can also use a context directly as a channel and assign it to a task or event.\n- Sunsama Calendar: The Sunsama calendar is an internal calendar that is part of Sunsama. Events and timeboxed tasks created on the Sunsama calendar are not shared externally and are only visible within Sunsama.\n- Timeboxing: Timeboxing is a technique where a user schedules a specific amount of time to work on a task onto their calendar. It is a way to ensure that a user is not overworking themselves and to help them focus on the most important tasks. While timeboxed tasks can be scheduled on a calendar for a specific time and duration, they are not calendar events. Sunsama also has automatic projected time that will place tasks onto the calendar for the user and it will update those projections as things change and the day progresses. Users can pin tasks to the calendar to override the projections.\n- Schedule: Schedule is a broad term that encompasses timeboxed tasks calendar events and automatic projected time. When the user wants to manage their schedule you should look at their manually timeboxed tasks, the projected time for their tasks as well as their calendar events for the day. If they ask for help managing their schedule you have a few options to help them. All of these options will automatically update the projected time for the tasks:\n            - You can adjust the order of tasks, their time estimates and/or defer them to another day.\n            - You can move calendar events to a new time on the calendar.\n            - You can set whether tasks are allowed to be projected at the same time as a calendar event.\n            - You can timebox tasks to the calendar manually.\n- Objective: An objective is a goal that a user can set for the week. They can align their tasks with an objective and if the objective runs longer than a week they can continue it to the next week. Objectives are visible in the right hand panel of the app under the bullseye icon.\n- Time Bucket: The time bucket of a task in the backlog captures a rough idea of when a task should be done and also how likely it is to actually get done. Options: "in the next two weeks", "in the next month", "in the next quarter", "in the next year", "someday", "never". \n\n\n\nSunsama FAQ:\n- How can I use Sunny in the app?: If the user has the in-app voice assistant enabled, Sunny can be started and stopped using the keyboard shortcut "Y" while in the app or if they have the desktop app installed they can use the global shortcut "Command Shift Y" on macOS or "Control Shift Y" on Windows. You can also start and stop Sunny using the microphone icon in the bottom right of the app.\n- How can I use Sunny over the phone?: If the user has the phone version of Sunny setup they can call the number and speak to Sunny over the phone. You can find the phone number in the Sunsama settings. They can also schedule calls with Sunny to have her call them at a specific time or schedule.\n- What can you do?: Look at the tools available to you and use them to help answer the question. Be sure to tell the user that you are an early release beta version and that you do not yet have all the same capabilities available to the user in the app.\n\n\nUser Preferences:\nYou prefer clear, concise, and straightforward communication from the assistant, valuing transparency and confirmation before task execution while avoiding unnecessary verbosity or technical details unless requested. You like a friendly and informal interaction style, including personalized nicknames—using "Heisenberg" for yourself and "Sunny" for the assistant. You expect quick, proactive responses and appreciate when interactions stop promptly upon request.\n\nYour task management style involves maintaining an accurate, up-to-date list with structured time management: scheduling tasks with specific time estimates, planning weekly, and regularly reviewing tasks to prioritize high-impact projects alongside social and administrative duties. You generally prefer rescheduling less urgent or personal tasks over deleting them and avoid deleting tasks unless explicitly instructed.\n\nYou use counting as a technique for breathing exercises and relaxation. You also keep certain important tasks (e.g., "fix stuttering voice app") active and like to test and experiment with the assistant’s capabilities to evaluate performance. Your primary calendar for events is Google Calendar.\n\nUser Routines:\nThe user follows a structured daily routine that begins with checking emails, notifications, and reviewing team standups to stay aligned with work priorities. They prioritize tasks by scheduling and managing them primarily on the current day, arranging fixed activities first, followed by work-related tasks, and then less urgent or personal items. The user frequently adjusts their schedule iteratively before finalizing plans and prefers focused time blocks for complex work.\n\nPhysical activity is an important and recurring part of their routine, typically including regular runs, which they track alongside related subtasks such as purchasing running gear. Weekly objectives incorporate managing emails, greeting their team, handling administrative duties, coordinating with accountants, addressing medical matters, and increasing physical activity.\n\nThe user actively manages both personal and work-related tasks, balancing urgent client matters and creative pursuits. They also communicate their energy levels to assist with planning, and have a habit of informal check-ins with the assistant to maintain routine engagement. Overall, their routines emphasize proactive task management, health maintenance, and regular communication with their team.\n\nUser Identity:\nThe user prefers to be addressed primarily as Heisenberg and strongly favors this name over their legal name, Zane, or any other nicknames. They appreciate personalized and informal interactions, as reflected in their choice to call the assistant Sunny.\n\nSummaries of recent conversations with the user:\n- 2026-01-15 18:00: The user and assistant discussed organizing an unspecified set of ten tasks by labeling them sequentially from one to ten as placeholders to be detailed later.\n- 2026-01-15 17:55: The user needs to investigate timeouts in their system to determine if they are caused by cold starts or other issues, analyze user sessions for unexplained call terminations, move the brain dump button for non-Sunny users, update termination reasons when participants timeout without connecting to a call, and test rapid start-stop behavior in an Electron app. These tasks were added to their backlog for the next two weeks.\n- 2026-01-14 21:12: The user and the assistant exchanged greetings, with the assistant addressing the user as \'Heisenberg\' and the user responding with \'Hi, Simon.\'\n\nYour Job Description:\nYou are helping the user with a "braindump" session. The user will ramble ideas, thoughts, tasks, etc. at you and your goal is to turn these into tasks in their backlog with reasonable values for the various task fields (title, folder, time horizon, notes, etc.).\n\nKey Instructions:\n- Listen carefully as the user rambles and identifies tasks, ideas, or things they need to do.\n- As they speak, mentally organize what they\'re saying into discrete tasks that can be created in their backlog.\n- Don\'t interrupt them while they\'re rambling - let them get everything out.\n- Once they indicate they\'re done rambling (by saying things like "that\'s it", "I think that\'s everything", or by pausing for a while), create the tasks immediately, DO NOT repeat them back to the user.\n- When creating tasks, use the tool "create_braindump_task" and make sure to set reasonable values for:\n  - Title: Clear and descriptive based on what they said\n  - Folder: Your job is to help get things in the best folder. Choose an appropriate folder based on the task and their preferences. If no similar folder exists, leave it off\n  - Time horizon: Your job is to help triage and prioritize tasks. Based on the actual task name, make a smart decision on the time horizon.\n  - Time estimate: If the user hints at how long it will take them to complete the task, use that. If not, leave it blank.\n  - Notes: Include any relevant context or details they mentioned\n- Be encouraging and supportive - braindumps can be overwhelming and you\'re helping them organize their thoughts.\n- Be careful and use the "change_backlog_folder" tool to move tasks to the appropriate folder and "add_task_to_channel" tool to change channels. Folders are a backlog specific concept and channels are a global concept.\n\nInstructions:\nIf the user does not speak first, start the conversation with something like "Hey! Just start rambling and I\'ll take care of organizing your thoughts into tasks." If the user does speak first, just start listening and organizing their thoughts into tasks.\n\nSome information is pre-fetched for you below. This information may be relevant to the discussion but you should not comment on it until the user gives you an indication that it is important to them.\n\nassistant :\n[{\'id\': \'6a196f03-ef63-41de-b98d-e1ff69813bea\', \'function\': {\'name\': \'get_user_info\', \'arguments\': \'{}\'}, \'type\': \'function\'}]\n\ntool - 6a196f03-ef63-41de-b98d-e1ff69813bea:\n{"user":{"name":"Zane Mccaig","email":"zane@sunsama.com","timezone":"America/Vancouver","maxWorkHoursPerDay":8,"currentTimeForUser":"11:04 AM","currentDayForUser":"2026-01-16","currentWeekDayForUser":"Friday","calendars":[{"id":"sunsama-7de256a2-e753-401f-b601-e5c05638dc97","name":"Sunsama Calendar","accessRole":"owner","provider":"sunsama-calendar","isDefaultCalendarForSchedulingEvents":false,"isDefaultCalendarForTimeboxing":true,"includedInAutoImportingOfEvents":false},{"id":"zane@sunsama.com","name":"Zane Mccaig","accessRole":"owner","provider":"google-calendar","isDefaultCalendarForSchedulingEvents":true,"isDefaultCalendarForTimeboxing":false,"includedInAutoImportingOfEvents":true},{"id":"c_ukuoebbn11gdv6n1v0jt18h89k@group.calendar.google.com","name":"test 1","accessRole":"owner","provider":"google-calendar","isDefaultCalendarForSchedulingEvents":false,"isDefaultCalendarForTimeboxing":false,"includedInAutoImportingOfEvents":false},{"id":"c_c71dvjrar7se12ekfqepph067c@group.calendar.google.com","name":"test 2","accessRole":"owner","provider":"google-calendar","isDefaultCalendarForSchedulingEvents":false,"isDefaultCalendarForTimeboxing":false,"includedInAutoImportingOfEvents":false},{"id":"c_h9uo876jh8dp3gaa2a3rglmjss@group.calendar.google.com","name":"test 3","accessRole":"owner","provider":"google-calendar","isDefaultCalendarForSchedulingEvents":false,"isDefaultCalendarForTimeboxing":false,"includedInAutoImportingOfEvents":false},{"id":"c_git731ot074kcnt4sq8cu4ji9s@group.calendar.google.com","name":"test 4","accessRole":"owner","provider":"google-calendar","isDefaultCalendarForSchedulingEvents":false,"isDefaultCalendarForTimeboxing":false,"includedInAutoImportingOfEvents":false},{"id":"c_ona71b8ovuv0deqrq67p16b2ms@group.calendar.google.com","name":"test 4","accessRole":"owner","provider":"google-calendar","isDefaultCalendarForSchedulingEvents":false,"isDefaultCalendarForTimeboxing":false,"includedInAutoImportingOfEvents":false},{"id":"c_eaq7nmm03upprf0j6e2akem2ms@group.calendar.google.com","name":"test 5","accessRole":"owner","provider":"google-calendar","isDefaultCalendarForSchedulingEvents":false,"isDefaultCalendarForTimeboxing":false,"includedInAutoImportingOfEvents":false},{"id":"sunsama.com_b9q30t893b2tp32bluq1t8rvg4@group.calendar.google.com","name":"Zane\'s personal test calendar","accessRole":"owner","provider":"google-calendar","isDefaultCalendarForSchedulingEvents":false,"isDefaultCalendarForTimeboxing":false,"includedInAutoImportingOfEvents":false},{"id":"en.canadian#holiday@group.v.calendar.google.com","name":"Holidays in Canada","accessRole":"reader","provider":"google-calendar","isDefaultCalendarForSchedulingEvents":false,"isDefaultCalendarForTimeboxing":false,"includedInAutoImportingOfEvents":false},{"id":"en.usa#holiday@group.v.calendar.google.com","name":"Holidays in United States","accessRole":"reader","provider":"google-calendar","isDefaultCalendarForSchedulingEvents":false,"isDefaultCalendarForTimeboxing":false,"includedInAutoImportingOfEvents":false},{"id":"ht3jlfaac5lfd6263ulfh4tql8@group.calendar.google.com","name":"Phases of the Moon","accessRole":"reader","provider":"google-calendar","isDefaultCalendarForSchedulingEvents":false,"isDefaultCalendarForTimeboxing":false,"includedInAutoImportingOfEvents":false},{"id":"s4ipng85dbia0t83fp316vflfq8rc7rm@import.calendar.google.com","name":"Zane’s tasks - sunsama.com (via Asana)","accessRole":"reader","provider":"google-calendar","isDefaultCalendarForSchedulingEvents":false,"isDefaultCalendarForTimeboxing":false,"includedInAutoImportingOfEvents":false},{"id":"AAkALgAAAAAAHYQDEapmEc2byACqAC-EWg0AKPy-EasNgkaYQCuEslBYeAAAAAANJQAA","name":"Calendar","accessRole":"owner","provider":"outlook-calendar","isDefaultCalendarForSchedulingEvents":false,"isDefaultCalendarForTimeboxing":false,"includedInAutoImportingOfEvents":true}],"callSchedule":[{"workflow":"PLANNING_CALL","schedule":[{"isoDayIndex":1,"dayName":"Monday","enabled":false,"time":null},{"isoDayIndex":2,"dayName":"Tuesday","enabled":true,"time":{"hour":9,"minute":30,"formatted":"9:30 AM"}},{"isoDayIndex":3,"dayName":"Wednesday","enabled":true,"time":{"hour":9,"minute":30,"formatted":"9:30 AM"}},{"isoDayIndex":4,"dayName":"Thursday","enabled":true,"time":{"hour":10,"minute":0,"formatted":"10:00 AM"}},{"isoDayIndex":5,"dayName":"Friday","enabled":false,"time":null},{"isoDayIndex":6,"dayName":"Saturday","enabled":false,"time":null},{"isoDayIndex":7,"dayName":"Sunday","enabled":false,"time":null}],"nextScheduledCall":"2026-01-14T17:30:00.000Z"}],"autoImportEventsEnabled":false,"importEventExclusionFilters":["hold","multi-day-all-day","single-day-all-day"],"autoImportEventCalendarIds":["zane@sunsama.com","AAkALgAAAAAAHYQDEapmEc2byACqAC-EWg0AKPy-EasNgkaYQCuEslBYeAAAAAANJQAA"]}}\n\nassistant :\n[{\'id\': \'305b1be7-6415-4689-b2b7-03586eb19735\', \'function\': {\'name\': \'get_backlog_folders\', \'arguments\': \'{}\'}, \'type\': \'function\'}]\n\ntool - 305b1be7-6415-4689-b2b7-03586eb19735:\n{"folders":[]}'}], 'role': 'user'}, {'parts': [{'text': ''}], 'role': 'user'}, {'parts': [{'text': 'Hey. Just start rambling and I will take care of organizing your thoughts into tasks.'}], 'role': 'model'}, {'parts': [{'text': "Sure. I'm just trying to fill up my day. I need five tasks labeled one through five. I'll fill them in later."}], 'role': 'user'}, {'parts': [{'function_call': {'id': 'ab4037a7-6e5e-44c0-9b3b-e13632386f30', 'args': {'timeBucket': 'someday', 'title': 'Task 1'}, 'name': 'create_braindump_task'}}], 'role': 'model'}, {'parts': [{'function_response': {'id': 'ab4037a7-6e5e-44c0-9b3b-e13632386f30', 'name': 'create_braindump_task', 'response': {'value': '{"content": "{\\"success\\":true,\\"task\\":{\\"_id\\":\\"696a8bb94ba9abe9e3dbee98\\",\\"title\\":\\"Task 1\\",\\"notes\\":\\"\\",\\"timeEstimate\\":\\"15 minutes\\",\\"sortOrder\\":-1768590265139,\\"isPersonal\\":false,\\"isWork\\":true,\\"isPrivate\\":false,\\"isArchived\\":false,\\"completed\\":false,\\"isBacklogged\\":true,\\"scheduledDate\\":\\"not scheduled for a specific date\\",\\"subtasks\\":[],\\"channel\\":\\"work\\",\\"folder\\":null,\\"timeboxEventIds\\":[]}}"}'}}}], 'role': 'user'}, {'parts': [{'function_call': {'id': '4ee49c06-29da-47cc-a51f-9eb903d5cbee', 'args': {'timeBucket': 'someday', 'title': 'Task 2'}, 'name': 'create_braindump_task'}}], 'role': 'model'}, {'parts': [{'function_response': {'id': '4ee49c06-29da-47cc-a51f-9eb903d5cbee', 'name': 'create_braindump_task', 'response': {'value': '{"content": "{\\"success\\":true,\\"task\\":{\\"_id\\":\\"696a8bb94ba9abe9e3dbee9c\\",\\"title\\":\\"Task 2\\",\\"notes\\":\\"\\",\\"timeEstimate\\":\\"15 minutes\\",\\"sortOrder\\":-1768590265140,\\"isPersonal\\":false,\\"isWork\\":true,\\"isPrivate\\":false,\\"isArchived\\":false,\\"completed\\":false,\\"isBacklogged\\":true,\\"scheduledDate\\":\\"not scheduled for a specific date\\",\\"subtasks\\":[],\\"channel\\":\\"work\\",\\"folder\\":null,\\"timeboxEventIds\\":[]}}"}'}}}], 'role': 'user'}, {'parts': [{'function_call': {'id': '0bfad4e3-e326-466b-b46e-d7a3a6cabd1f', 'args': {'title': 'Task 3', 'timeBucket': 'someday'}, 'name': 'create_braindump_task'}}], 'role': 'model'}, {'parts': [{'function_response': {'id': '0bfad4e3-e326-466b-b46e-d7a3a6cabd1f', 'name': 'create_braindump_task', 'response': {'value': '{"content": "{\\"success\\":true,\\"task\\":{\\"_id\\":\\"696a8bb94ba9abe9e3dbee9e\\",\\"title\\":\\"Task 3\\",\\"notes\\":\\"\\",\\"timeEstimate\\":\\"15 minutes\\",\\"sortOrder\\":-1768590265140,\\"isPersonal\\":false,\\"isWork\\":true,\\"isPrivate\\":false,\\"isArchived\\":false,\\"completed\\":false,\\"isBacklogged\\":true,\\"scheduledDate\\":\\"not scheduled for a specific date\\",\\"subtasks\\":[],\\"channel\\":\\"work\\",\\"folder\\":null,\\"timeboxEventIds\\":[]}}"}'}}}], 'role': 'user'}, {'parts': [{'function_call': {'id': 'bdda1d96-7e9f-42e6-840a-0d29d17bc52a', 'args': {'timeBucket': 'someday', 'title': 'Task 4'}, 'name': 'create_braindump_task'}}], 'role': 'model'}, {'parts': [{'function_response': {'id': 'bdda1d96-7e9f-42e6-840a-0d29d17bc52a', 'name': 'create_braindump_task', 'response': {'value': '{"content": "{\\"success\\":true,\\"task\\":{\\"_id\\":\\"696a8bb94ba9abe9e3dbee9a\\",\\"title\\":\\"Task 4\\",\\"notes\\":\\"\\",\\"timeEstimate\\":\\"15 minutes\\",\\"sortOrder\\":-1768590265140,\\"isPersonal\\":false,\\"isWork\\":true,\\"isPrivate\\":false,\\"isArchived\\":false,\\"completed\\":false,\\"isBacklogged\\":true,\\"scheduledDate\\":\\"not scheduled for a specific date\\",\\"subtasks\\":[],\\"channel\\":\\"work\\",\\"folder\\":null,\\"timeboxEventIds\\":[]}}"}'}}}], 'role': 'user'}, {'parts': [{'function_call': {'id': 'db77c745-7f1a-4644-bc24-7cf5fd728774', 'args': {'title': 'Task 5', 'timeBucket': 'someday'}, 'name': 'create_braindump_task'}}], 'role': 'model'}, {'parts': [{'function_response': {'id': 'db77c745-7f1a-4644-bc24-7cf5fd728774', 'name': 'create_braindump_task', 'response': {'value': '{"content": "{\\"success\\":true,\\"task\\":{\\"_id\\":\\"696a8bb94ba9abe9e3dbeea0\\",\\"title\\":\\"Task 5\\",\\"notes\\":\\"\\",\\"timeEstimate\\":\\"15 minutes\\",\\"sortOrder\\":-1768590265140,\\"isPersonal\\":false,\\"isWork\\":true,\\"isPrivate\\":false,\\"isArchived\\":false,\\"completed\\":false,\\"isBacklogged\\":true,\\"scheduledDate\\":\\"not scheduled for a specific date\\",\\"subtasks\\":[],\\"channel\\":\\"work\\",\\"folder\\":null,\\"timeboxEventIds\\":[]}}"}'}}}], 'role': 'user'}, {'parts': [{'function_call': {'args': {'timeBucket': 'someday', 'title': 'Task 2'}, 'name': 'create_braindump_task'}}], 'role': 'model'}, {'parts': [{'function_response': {'name': 'tool_call_result', 'response': {'content': '{"success":true,"task":{"_id":"696a8bb94ba9abe9e3dbee9c","title":"Task 2","notes":"","timeEstimate":"15 minutes","sortOrder":-1768590265140,"isPersonal":false,"isWork":true,"isPrivate":false,"isArchived":false,"completed":false,"isBacklogged":true,"scheduledDate":"not scheduled for a specific date","subtasks":[],"channel":"work","folder":null,"timeboxEventIds":[]}}'}}}], 'role': 'model'}, {'parts': [{'function_call': {'args': {'timeBucket': 'someday', 'title': 'Task 4'}, 'name': 'create_braindump_task'}}], 'role': 'model'}, {'parts': [{'function_response': {'name': 'tool_call_result', 'response': {'content': '{"success":true,"task":{"_id":"696a8bb94ba9abe9e3dbee9a","title":"Task 4","notes":"","timeEstimate":"15 minutes","sortOrder":-1768590265140,"isPersonal":false,"isWork":true,"isPrivate":false,"isArchived":false,"completed":false,"isBacklogged":true,"scheduledDate":"not scheduled for a specific date","subtasks":[],"channel":"work","folder":null,"timeboxEventIds":[]}}'}}}], 'role': 'model'}, {'parts': [{'function_call': {'args': {'timeBucket': 'someday', 'title': 'Task 1'}, 'name': 'create_braindump_task'}}], 'role': 'model'}, {'parts': [{'function_response': {'name': 'tool_call_result', 'response': {'content': '{"success":true,"task":{"_id":"696a8bb94ba9abe9e3dbee98","title":"Task 1","notes":"","timeEstimate":"15 minutes","sortOrder":-1768590265139,"isPersonal":false,"isWork":true,"isPrivate":false,"isArchived":false,"completed":false,"isBacklogged":true,"scheduledDate":"not scheduled for a specific date","subtasks":[],"channel":"work","folder":null,"timeboxEventIds":[]}}'}}}], 'role': 'model'}, {'parts': [{'function_call': {'args': {'title': 'Task 3', 'timeBucket': 'someday'}, 'name': 'create_braindump_task'}}], 'role': 'model'}, {'parts': [{'function_response': {'name': 'tool_call_result', 'response': {'content': '{"success":true,"task":{"_id":"696a8bb94ba9abe9e3dbee9e","title":"Task 3","notes":"","timeEstimate":"15 minutes","sortOrder":-1768590265140,"isPersonal":false,"isWork":true,"isPrivate":false,"isArchived":false,"completed":false,"isBacklogged":true,"scheduledDate":"not scheduled for a specific date","subtasks":[],"channel":"work","folder":null,"timeboxEventIds":[]}}'}}}], 'role': 'model'}, {'parts': [{'function_call': {'args': {'title': 'Task 5', 'timeBucket': 'someday'}, 'name': 'create_braindump_task'}}], 'role': 'model'}, {'parts': [{'function_response': {'name': 'tool_call_result', 'response': {'content': 'Task already exists.'}}}], 'role': 'model'}, {'parts': [{'text': 'Got it, Heisenberg. I have added those five placeholder tasks to your backlog. Is there anything else I can help you with for now?'}], 'role': 'model'}, {'parts': [{'function_call': {'id': '6d6bce49-5c24-4c6e-b1b4-d042fd24a72f', 'args': {'timeBucket': 'someday', 'title': 'Task 5'}, 'name': 'create_braindump_task'}}], 'role': 'model'}, {'parts': [{'function_response': {'id': '6d6bce49-5c24-4c6e-b1b4-d042fd24a72f', 'name': 'create_braindump_task', 'response': {'value': '{"content": "Task already exists."}'}}}], 'role': 'user'}, {'parts': [{'function_call': {'id': '567b1b3c-0871-4dc8-9f34-fe8592d2c922', 'args': {'title': 'Task 5', 'timeBucket': 'someday'}, 'name': 'create_braindump_task'}}], 'role': 'model'}, {'parts': [{'function_response': {'id': '567b1b3c-0871-4dc8-9f34-fe8592d2c922', 'name': 'create_braindump_task', 'response': {'value': '{"content": "Task already exists."}'}}}], 'role': 'user'}, {'parts': [{'function_call': {'id': '363c5296-e80d-45a1-9dc4-ece336bbb77a', 'args': {'timeBucket': 'someday', 'title': 'Task 5'}, 'name': 'create_braindump_task'}}], 'role': 'model'}, {'parts': [{'function_response': {'id': '363c5296-e80d-45a1-9dc4-ece336bbb77a', 'name': 'create_braindump_task', 'response': {'value': '{"content": "Task already exists."}'}}}], 'role': 'user'}, {'parts': [{'text': 'Got it. I have added five tasks to your backlog, labeled one through five. Is there anything else I can help you with?'}], 'role': 'model'}, {'parts': [{'text': 'Are you still there?'}], 'role': 'model'}, {'parts': [{'text': 'Now you can go.'}], 'role': 'user'}, {'parts': [{'function_call': {'id': '2522e212-ad57-41de-901e-d535655693c6', 'args': {}, 'name': 'terminate_call'}}], 'role': 'model'}, {'parts': [{'function_response': {'id': '2522e212-ad57-41de-901e-d535655693c6', 'name': 'terminate_call', 'response': {'value': '{"content": "{\'status\': \'completed\'}"}'}}}], 'role': 'user'}]
-[voice] DEBUG:pipecat.services.google.llm:System instruction changed: Thank the customer for their time and end the conversation.
-[voice] INFO:google_genai.models:AFC is enabled with max remote calls: 10.
-[voice] INFO:sunsama_voice.bot_utils:Sent server message to client: type=processing_update, data={'status': 'request_sent'}
-[voice] DEBUG:pipecat.processors.metrics.frame_processor_metrics:GoogleLLMService#0 TTFB: 0.659785270690918
-[voice] INFO:sunsama_voice.bot_utils:Sent server message to client: type=processing_update, data={'status': 'first_byte_received'}
-[voice] INFO:sunsama_voice.bot_utils:Sent server message to client: type=processing_update, data={'status': 'first_byte_received'}
-[voice] DEBUG:pipecat.services.cartesia.tts:CartesiaTTSService#1: Generating TTS [Alright, Heisenberg.]
-[voice] DEBUG:pipecat.processors.metrics.frame_processor_metrics:CartesiaTTSService#1 usage characters: 20
-[voice] DEBUG:pipecat.processors.metrics.frame_processor_metrics:CartesiaTTSService#1 processing time: 0.0008521080017089844
-[voice] DEBUG:pipecat.processors.metrics.frame_processor_metrics:GoogleLLMService#0 prompt tokens: 7890, completion tokens: 10
-[voice] INFO:sunsama_voice.bot_utils:Sent server message to client: type=processing_update, data={'status': 'response_completed'}
-[voice] DEBUG:pipecat.services.cartesia.tts:CartesiaTTSService#1: Generating TTS [Thank you for your time.]
-[voice] DEBUG:pipecat.processors.metrics.frame_processor_metrics:CartesiaTTSService#1 usage characters: 24
-[voice] DEBUG:pipecat.processors.metrics.frame_processor_metrics:CartesiaTTSService#1 processing time: 0.0007202625274658203
-[voice] DEBUG:pipecat.processors.metrics.frame_processor_metrics:CartesiaTTSService#1 TTFB: 0.19838976860046387
-[voice] DEBUG:pipecat.transports.base_output:Bot started speaking
-[voice] <string>:4: DeprecationWarning: OpenAILLMContextFrame is deprecated and will be removed in a future version. Use LLMContextFrame with the universal `LLMContext` and `LLMContextAggregatorPair` instead. See OpenAILLMContext docstring for migration guide.
-[voice] <string>:4: DeprecationWarning: OpenAILLMContextAssistantTimestampFrame is deprecated and will be removed in a future version. Use LLMContextAssistantTimestampFrame with the universal LLMContext and LLMContextAggregatorPair instead. See OpenAILLMContext docstring for migration guide.
-[voice] DEBUG:pipecat.transports.base_output:Bot stopped speaking
-[voice] DEBUG:pipecat.services.cartesia.tts:Disconnecting from Cartesia
-[voice] INFO:pipecat.transports.daily.transport:Leaving https://cloud-99400c183d52498c940fe96f45c05b44.daily.co/teWpRxCQnkMHYi7Q3lqh
-[voice] DEBUG:pipecat.transports.daily.transport:Stopping transcription
-[voice] DEBUG:pipecat.transports.daily.transport:Transcription stopped
-[voice] INFO:pipecat.transports.daily.transport:Left https://cloud-99400c183d52498c940fe96f45c05b44.daily.co/teWpRxCQnkMHYi7Q3lqh
-[voice] DEBUG:pipecat.pipeline.task:PipelineTask#0: EndFrame#0(reason: None) reached the end of the pipeline, pipeline is closing.
-[voice] DEBUG:pipecat.pipeline.task:Pipeline task PipelineTask#0 is finishing...
-[voice] DEBUG:pipecat.pipeline.task:Pipeline task PipelineTask#0 has finished
-[voice] DEBUG:pipecat.pipeline.runner:Garbage collector: collected 112 objects.
-[voice] DEBUG:pipecat.pipeline.runner:Garbage collector: uncollectable objects []
-[voice] DEBUG:pipecat.pipeline.runner:Runner PipelineRunner#0 finished running PipelineTask#0
-[voice] INFO:sunsama_voice.bot:Timer: Pipeline completed in 81.08789300918579 seconds

From b6341ffaa59f518fdb70978d473fc8829524eeab Mon Sep 17 00:00:00 2001
From: marcus-daily <111281783+marcus-daily@users.noreply.github.com>
Date: Thu, 22 Jan 2026 13:24:30 +0000
Subject: [PATCH 43/51] Save Smart Turn input data if SMART_TURN_LOG_DATA is
 set

---
 .gitignore                                    | 11 +++-
 .../turn/smart_turn/local_smart_turn_v3.py    | 54 ++++++++++++++++++-
 src/pipecat/utils/env.py                      | 54 +++++++++++++++++++
 3 files changed, 116 insertions(+), 3 deletions(-)
 create mode 100644 src/pipecat/utils/env.py

diff --git a/.gitignore b/.gitignore
index 426f9be3a..512ec7316 100644
--- a/.gitignore
+++ b/.gitignore
@@ -4,7 +4,14 @@ __pycache__/
 *~
 venv
 .venv
-/.idea
+.idea
+.gradle
+.next
+next-env.d.ts
+local.properties
+*.log
+*.lock
+smart_turn_audio_log
 #*#
 
 # Distribution / Packaging
@@ -27,7 +34,7 @@ share/python-wheels/
 *.egg
 MANIFEST
 .DS_Store
-.env
+.env*
 fly.toml
 
 # Examples
diff --git a/src/pipecat/audio/turn/smart_turn/local_smart_turn_v3.py b/src/pipecat/audio/turn/smart_turn/local_smart_turn_v3.py
index 0907ab28f..e62131e0c 100644
--- a/src/pipecat/audio/turn/smart_turn/local_smart_turn_v3.py
+++ b/src/pipecat/audio/turn/smart_turn/local_smart_turn_v3.py
@@ -16,6 +16,7 @@ import numpy as np
 from loguru import logger
 
 from pipecat.audio.turn.smart_turn.base_smart_turn import BaseSmartTurn
+from pipecat.utils.env import env_truthy
 
 try:
     import onnxruntime as ort
@@ -48,6 +49,8 @@ class LocalSmartTurnAnalyzerV3(BaseSmartTurn):
         """
         super().__init__(**kwargs)
 
+        self._log_data = env_truthy("SMART_TURN_LOG_DATA", default=False)
+
         if not smart_turn_model_path:
             # Load bundled model
             model_name = "smart-turn-v3.2-cpu.onnx"
@@ -81,6 +84,49 @@ class LocalSmartTurnAnalyzerV3(BaseSmartTurn):
 
         logger.debug("Loaded Local Smart Turn v3.x")
 
+    def _write_audio_to_wav(
+        self, audio_array: np.ndarray, sample_rate: int = 16000, suffix: str = ""
+    ) -> None:
+        """Write audio data to a WAV file in a background thread.
+
+        Args:
+            audio_array: The audio data as a numpy array (float32, normalized to [-1, 1]).
+            sample_rate: The sample rate of the audio data.
+            suffix: Optional suffix to append to the filename (e.g., "_raw", "_padded").
+        """
+        import wave
+        from datetime import datetime
+        import os
+        import threading
+
+        # Generate filename with current timestamp (millisecond precision)
+        timestamp = datetime.now().strftime("%Y-%m-%d__%H:%M:%S.%f")[:-3]
+        log_dir = "./smart_turn_audio_log"
+        os.makedirs(log_dir, exist_ok=True)
+        filename = os.path.join(log_dir, f"{timestamp}{suffix}.wav")
+
+        # Make a copy of the audio data to avoid issues with the array being modified
+        audio_copy = audio_array.copy()
+
+        def write_wav():
+            try:
+                # Convert float32 audio to int16 for WAV file
+                audio_int16 = (audio_copy * 32767).astype(np.int16)
+
+                with wave.open(filename, "wb") as wav_file:
+                    wav_file.setnchannels(1)  # Mono
+                    wav_file.setsampwidth(2)  # 2 bytes for int16
+                    wav_file.setframerate(sample_rate)
+                    wav_file.writeframes(audio_int16.tobytes())
+
+                logger.debug(f"Wrote audio to {filename}")
+            except Exception as e:
+                logger.error(f"Failed to write audio to {filename}: {e}")
+
+        # Start background thread to write the WAV file
+        thread = threading.Thread(target=write_wav, daemon=True)
+        thread.start()
+
     def _predict_endpoint(self, audio_array: np.ndarray) -> Dict[str, Any]:
         """Predict end-of-turn using local ONNX model."""
 
@@ -95,6 +141,8 @@ class LocalSmartTurnAnalyzerV3(BaseSmartTurn):
                 return np.pad(audio_array, (padding, 0), mode="constant", constant_values=0)
             return audio_array
 
+        audio_for_logging = audio_array
+
         # Truncate to 8 seconds (keeping the end) or pad to 8 seconds
         audio_array = truncate_audio_to_last_n_seconds(audio_array, n_seconds=8)
 
@@ -122,7 +170,11 @@ class LocalSmartTurnAnalyzerV3(BaseSmartTurn):
         # Make prediction (1 for Complete, 0 for Incomplete)
         prediction = 1 if probability > 0.5 else 0
 
+        if self._log_data:
+            suffix = "_complete" if prediction == 1 else "_incomplete"
+            self._write_audio_to_wav(audio_for_logging, sample_rate=16000, suffix=suffix)
+
         return {
             "prediction": prediction,
             "probability": probability,
-        }
+        }
\ No newline at end of file
diff --git a/src/pipecat/utils/env.py b/src/pipecat/utils/env.py
new file mode 100644
index 000000000..b3467c1f4
--- /dev/null
+++ b/src/pipecat/utils/env.py
@@ -0,0 +1,54 @@
+#
+# Copyright (c) 2024-2026, Daily
+#
+# SPDX-License-Identifier: BSD 2-Clause License
+#
+
+"""Environment variable helpers.
+
+This module provides small, centralized parsing helpers for environment variables.
+"""
+
+from __future__ import annotations
+
+import os
+
+
+class InvalidEnvVarValueError(ValueError):
+    """Raised when an environment variable value cannot be parsed."""
+
+    def __init__(self, name: str, value: str, expected: str):
+        super().__init__(f"Invalid value for env var {name!r}: {value!r}. Expected {expected}.")
+        self.name = name
+        self.value = value
+        self.expected = expected
+
+
+def env_truthy(name: str, default: bool = False) -> bool:
+    """Interpret an environment variable as a boolean.
+
+    - If the variable is **not set**, returns `default`.
+    - If the variable is set to a recognized boolean string, returns the parsed value.
+    - Otherwise, raises `InvalidEnvVarValueError`.
+
+    Recognized values (case-insensitive, whitespace ignored):
+    - Truthy: "1", "true", "yes", "y", "on"
+    - Falsy:  "0", "false", "no", "n", "off", ""
+    """
+
+    raw = os.getenv(name)
+    if raw is None:
+        return default
+
+    val = raw.strip().lower()
+    if val in {"1", "true", "yes", "y", "on"}:
+        return True
+    if val in {"0", "false", "no", "n", "off", ""}:
+        return False
+
+    raise InvalidEnvVarValueError(
+        name=name,
+        value=raw,
+        expected='true or false',
+    )
+

From 76b774072c429065b3d35909ad73e94e1a292e95 Mon Sep 17 00:00:00 2001
From: marcus-daily <111281783+marcus-daily@users.noreply.github.com>
Date: Thu, 22 Jan 2026 13:34:52 +0000
Subject: [PATCH 44/51] Formatting fixes

---
 src/pipecat/audio/turn/smart_turn/local_smart_turn_v3.py | 6 +++---
 src/pipecat/utils/env.py                                 | 5 ++---
 2 files changed, 5 insertions(+), 6 deletions(-)

diff --git a/src/pipecat/audio/turn/smart_turn/local_smart_turn_v3.py b/src/pipecat/audio/turn/smart_turn/local_smart_turn_v3.py
index e62131e0c..84f6b5622 100644
--- a/src/pipecat/audio/turn/smart_turn/local_smart_turn_v3.py
+++ b/src/pipecat/audio/turn/smart_turn/local_smart_turn_v3.py
@@ -94,10 +94,10 @@ class LocalSmartTurnAnalyzerV3(BaseSmartTurn):
             sample_rate: The sample rate of the audio data.
             suffix: Optional suffix to append to the filename (e.g., "_raw", "_padded").
         """
-        import wave
-        from datetime import datetime
         import os
         import threading
+        import wave
+        from datetime import datetime
 
         # Generate filename with current timestamp (millisecond precision)
         timestamp = datetime.now().strftime("%Y-%m-%d__%H:%M:%S.%f")[:-3]
@@ -177,4 +177,4 @@ class LocalSmartTurnAnalyzerV3(BaseSmartTurn):
         return {
             "prediction": prediction,
             "probability": probability,
-        }
\ No newline at end of file
+        }
diff --git a/src/pipecat/utils/env.py b/src/pipecat/utils/env.py
index b3467c1f4..cb66a5400 100644
--- a/src/pipecat/utils/env.py
+++ b/src/pipecat/utils/env.py
@@ -18,6 +18,7 @@ class InvalidEnvVarValueError(ValueError):
     """Raised when an environment variable value cannot be parsed."""
 
     def __init__(self, name: str, value: str, expected: str):
+        """Initialize an InvalidEnvVarValueError."""
         super().__init__(f"Invalid value for env var {name!r}: {value!r}. Expected {expected}.")
         self.name = name
         self.value = value
@@ -35,7 +36,6 @@ def env_truthy(name: str, default: bool = False) -> bool:
     - Truthy: "1", "true", "yes", "y", "on"
     - Falsy:  "0", "false", "no", "n", "off", ""
     """
-
     raw = os.getenv(name)
     if raw is None:
         return default
@@ -49,6 +49,5 @@ def env_truthy(name: str, default: bool = False) -> bool:
     raise InvalidEnvVarValueError(
         name=name,
         value=raw,
-        expected='true or false',
+        expected="true or false",
     )
-

From f2fa5d9733178087374112ab117fc13542a281ea Mon Sep 17 00:00:00 2001
From: marcus-daily <111281783+marcus-daily@users.noreply.github.com>
Date: Thu, 22 Jan 2026 14:16:44 +0000
Subject: [PATCH 45/51] Updating changelog

---
 changelog/3525.added.md | 1 +
 1 file changed, 1 insertion(+)
 create mode 100644 changelog/3525.added.md

diff --git a/changelog/3525.added.md b/changelog/3525.added.md
new file mode 100644
index 000000000..0dcf74287
--- /dev/null
+++ b/changelog/3525.added.md
@@ -0,0 +1 @@
+- Added new `SMART_TURN_LOG_DATA` environment variable, which causes Smart Turn input data to be saved to disk

From b13b65d6e295e981f8e3818d66becbc728b69c3b Mon Sep 17 00:00:00 2001
From: Waldek Maleska <w.maleska@gmail.com>
Date: Thu, 22 Jan 2026 15:17:41 +0000
Subject: [PATCH 46/51] Update README.md - fix Google Imagen URL

---
 README.md | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/README.md b/README.md
index 65a12634f..e3ebaba09 100644
--- a/README.md
+++ b/README.md
@@ -81,7 +81,7 @@ Catch new features, interviews, and how-tos on our [Pipecat TV](https://www.yout
 | Serializers         | [Exotel](https://docs.pipecat.ai/server/utilities/serializers/exotel), [Plivo](https://docs.pipecat.ai/server/utilities/serializers/plivo), [Twilio](https://docs.pipecat.ai/server/utilities/serializers/twilio), [Telnyx](https://docs.pipecat.ai/server/utilities/serializers/telnyx), [Vonage](https://docs.pipecat.ai/server/utilities/serializers/vonage)                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                           |
 | Video               | [HeyGen](https://docs.pipecat.ai/server/services/video/heygen), [Tavus](https://docs.pipecat.ai/server/services/video/tavus), [Simli](https://docs.pipecat.ai/server/services/video/simli)                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                |
 | Memory              | [mem0](https://docs.pipecat.ai/server/services/memory/mem0)                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                               |
-| Vision & Image      | [fal](https://docs.pipecat.ai/server/services/image-generation/fal), [Google Imagen](https://docs.pipecat.ai/server/services/image-generation/fal), [Moondream](https://docs.pipecat.ai/server/services/vision/moondream)                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                 |
+| Vision & Image      | [fal](https://docs.pipecat.ai/server/services/image-generation/fal), [Google Imagen](https://docs.pipecat.ai/server/services/image-generation/google-imagen), [Moondream](https://docs.pipecat.ai/server/services/vision/moondream)                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                 |
 | Audio Processing    | [Silero VAD](https://docs.pipecat.ai/server/utilities/audio/silero-vad-analyzer), [Krisp](https://docs.pipecat.ai/server/utilities/audio/krisp-filter), [Koala](https://docs.pipecat.ai/server/utilities/audio/koala-filter), [ai-coustics](https://docs.pipecat.ai/server/utilities/audio/aic-filter)                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                    |
 | Analytics & Metrics | [OpenTelemetry](https://docs.pipecat.ai/server/utilities/opentelemetry), [Sentry](https://docs.pipecat.ai/server/services/analytics/sentry)                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                               |
 

From 6b5bcae86fe9d524e1f47531681b1469c93c0394 Mon Sep 17 00:00:00 2001
From: Cale Shapera <25466659+cshape@users.noreply.github.com>
Date: Thu, 22 Jan 2026 11:21:15 -0800
Subject: [PATCH 47/51] change default Inworld TTS model to inworld-tts-1.5-max
 (#3531)

---
 changelog/3531.changed.md           | 2 ++
 src/pipecat/services/inworld/tts.py | 4 ++--
 2 files changed, 4 insertions(+), 2 deletions(-)
 create mode 100644 changelog/3531.changed.md

diff --git a/changelog/3531.changed.md b/changelog/3531.changed.md
new file mode 100644
index 000000000..326d8d4d1
--- /dev/null
+++ b/changelog/3531.changed.md
@@ -0,0 +1,2 @@
+- Changed default Inworld TTS model from `inworld-tts-1` to                      
+ `inworld-tts-1.5-max`.
diff --git a/src/pipecat/services/inworld/tts.py b/src/pipecat/services/inworld/tts.py
index 38617ac15..4e6ec3f4e 100644
--- a/src/pipecat/services/inworld/tts.py
+++ b/src/pipecat/services/inworld/tts.py
@@ -72,7 +72,7 @@ class InworldHttpTTSService(WordTTSService):
         api_key: str,
         aiohttp_session: aiohttp.ClientSession,
         voice_id: str = "Ashley",
-        model: str = "inworld-tts-1",
+        model: str = "inworld-tts-1.5-max",
         streaming: bool = True,
         sample_rate: Optional[int] = None,
         encoding: str = "LINEAR16",
@@ -427,7 +427,7 @@ class InworldTTSService(AudioContextWordTTSService):
         *,
         api_key: str,
         voice_id: str = "Ashley",
-        model: str = "inworld-tts-1",
+        model: str = "inworld-tts-1.5-max",
         url: str = "wss://api.inworld.ai/tts/v1/voice:streamBidirectional",
         sample_rate: Optional[int] = None,
         encoding: str = "LINEAR16",

From 82a799e63e4bc296b97cfbab041f61917150be40 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Aleix=20Conchillo=20Flaqu=C3=A9?= <aleix@daily.co>
Date: Thu, 22 Jan 2026 12:53:38 -0800
Subject: [PATCH 48/51] claude: add docstring skill

---
 .claude/skills/docstring/SKILL.md | 257 ++++++++++++++++++++++++++++++
 1 file changed, 257 insertions(+)
 create mode 100644 .claude/skills/docstring/SKILL.md

diff --git a/.claude/skills/docstring/SKILL.md b/.claude/skills/docstring/SKILL.md
new file mode 100644
index 000000000..1c1e3c905
--- /dev/null
+++ b/.claude/skills/docstring/SKILL.md
@@ -0,0 +1,257 @@
+---
+name: docstring
+description: Document a Python module and its classes using Google style
+---
+
+Document a Python module and its classes using Google-style docstrings following project conventions. The class name is provided as an argument.
+
+## Instructions
+
+1. First, find the class in the codebase:
+   ```
+   Search for "class ClassName" in src/pipecat/
+   ```
+
+2. If multiple files contain that class name:
+   - List all matches with their file paths
+   - Ask the user which one they want to document
+   - Wait for confirmation before proceeding
+
+3. Once the file is identified, read the module to understand its structure:
+   - Identify all classes, functions, and important type aliases
+   - Understand the purpose of each component
+
+4. Apply documentation in this order:
+   - Module docstring (at top, after imports)
+   - Class docstrings
+   - `__init__` methods (always document constructor parameters)
+   - Public methods (not starting with `_`)
+   - Dataclass/config classes with field descriptions
+
+5. Skip documentation for:
+   - Private methods (starting with `_`)
+   - Simple dunder methods (`__str__`, `__repr__`, `__post_init__`)
+   - Very simple pass-through properties
+   - **Already documented code** - If a class, method, or function already has a complete docstring that follows the project style, do not modify it. A docstring is complete if it has:
+     - A one-line summary
+     - Args section (if it has parameters)
+     - Returns section (if it returns something meaningful)
+   - Only add or improve documentation where it is missing or incomplete
+
+## Module Docstring Format
+
+```python
+"""[One-line description of module purpose].
+
+[Optional: Longer explanation of functionality, key classes, or use cases.]
+"""
+```
+
+Example:
+```python
+"""Neuphonic text-to-speech service implementations.
+
+This module provides WebSocket and HTTP-based integrations with Neuphonic's
+text-to-speech API for real-time audio synthesis.
+"""
+```
+
+## Class Docstring Format
+
+```python
+class ClassName:
+    """One-line summary describing what the class does.
+
+    [Longer description explaining purpose, behavior, and key features.
+    Use action-oriented language.]
+
+    [Optional: Event handlers, usage notes, or important caveats.]
+    """
+```
+
+Example:
+```python
+class FrameProcessor(BaseObject):
+    """Base class for all frame processors in the pipeline.
+
+    Frame processors are the building blocks of Pipecat pipelines, they can be
+    linked to form complex processing pipelines. They receive frames, process
+    them, and pass them to the next or previous processor in the chain.
+
+    Event handlers available:
+
+    - on_before_process_frame: Called before a frame is processed
+    - on_after_process_frame: Called after a frame is processed
+
+    Example::
+
+        @processor.event_handler("on_before_process_frame")
+        async def on_before_process_frame(processor, frame):
+            ...
+
+        @processor.event_handler("on_after_process_frame")
+        async def on_after_process_frame(processor, frame):
+            ...
+    """
+```
+
+Note: When listing event handlers, do NOT use backticks. Include an `Example::` section (with double colon for Sphinx) showing the decorator pattern and function signature for each event.
+
+## Constructor (`__init__`) Format
+
+```python
+def __init__(self, *, param1: Type, param2: Type = default, **kwargs):
+    """Initialize the [ClassName].
+
+    Args:
+        param1: Description of param1 and its purpose.
+        param2: Description of param2. Defaults to [default].
+        **kwargs: Additional arguments passed to parent class.
+    """
+```
+
+Example:
+```python
+def __init__(
+    self,
+    *,
+    api_key: str,
+    voice_id: Optional[str] = None,
+    sample_rate: Optional[int] = 22050,
+    **kwargs,
+):
+    """Initialize the Neuphonic TTS service.
+
+    Args:
+        api_key: Neuphonic API key for authentication.
+        voice_id: ID of the voice to use for synthesis.
+        sample_rate: Audio sample rate in Hz. Defaults to 22050.
+        **kwargs: Additional arguments passed to parent InterruptibleTTSService.
+    """
+```
+
+## Method Docstring Format
+
+```python
+async def method_name(self, param1: Type) -> ReturnType:
+    """One-line summary of what method does.
+
+    [Longer description if behavior isn't obvious.]
+
+    Args:
+        param1: Description of param1.
+
+    Returns:
+        Description of return value.
+
+    Raises:
+        ExceptionType: When this exception is raised.
+    """
+```
+
+Example:
+```python
+async def put(self, item: Tuple[Frame, FrameDirection, FrameCallback]):
+    """Put an item into the priority queue.
+
+    System frames (`SystemFrame`) have higher priority than any other
+    frames. If a non-frame item is provided it will have the highest priority.
+
+    Args:
+        item: The item to enqueue.
+    """
+```
+
+## Dataclass/Config Format
+
+```python
+@dataclass
+class ConfigName:
+    """One-line description of configuration.
+
+    [Explanation of when/how to use this config.]
+
+    Parameters:
+        field1: Description of field1.
+        field2: Description of field2. Defaults to [default].
+    """
+
+    field1: Type
+    field2: Type = default_value
+```
+
+Example:
+```python
+@dataclass
+class FrameProcessorSetup:
+    """Configuration parameters for frame processor initialization.
+
+    Parameters:
+        clock: The clock instance for timing operations.
+        task_manager: The task manager for handling async operations.
+        observer: Optional observer for monitoring frame processing events.
+    """
+
+    clock: BaseClock
+    task_manager: BaseTaskManager
+    observer: Optional[BaseObserver] = None
+```
+
+## Enum Documentation Format
+
+```python
+class EnumName(Enum):
+    """One-line description of the enum purpose.
+
+    [Longer description of how the enum is used.]
+
+    Parameters:
+        VALUE1: Description of VALUE1.
+        VALUE2: Description of VALUE2.
+    """
+
+    VALUE1 = 1
+    VALUE2 = 2
+```
+
+## Writing Style Guidelines
+
+- **Concise and professional** - No casual language or filler words
+- **Action-oriented** - Start with verbs: "Processes...", "Manages...", "Converts..."
+- **Purpose before implementation** - Explain WHY before HOW
+- **Clear parameter descriptions** - Include type hints, defaults, and purpose
+- **No redundant type info** - Type hints are in the signature, don't repeat in description
+- **Use backticks for code references** - Wrap class names, method names, event names, parameter names, and code snippets in backticks
+
+Good: "Neuphonic API key for authentication."
+Bad: "str: The API key (string) that is used for authenticating with Neuphonic."
+
+Good: "Triggers `on_speech_started` when the `VADAnalyzer` detects speech."
+Bad: "Triggers on_speech_started when the VADAnalyzer detects speech."
+
+## Deprecation Notice Format
+
+When documenting deprecated code:
+
+```python
+"""[Description].
+
+.. deprecated:: X.X.X
+    `ClassName` is deprecated and will be removed in a future version.
+    Use `NewClassName` instead.
+"""
+```
+
+## Checklist
+
+Before finishing, verify:
+
+- [ ] Module has a docstring at the top (after copyright header and imports)
+- [ ] All public classes have docstrings
+- [ ] All `__init__` methods document their parameters
+- [ ] All public methods have docstrings with Args/Returns/Raises as needed
+- [ ] Dataclasses use "Parameters:" section for field descriptions
+- [ ] Enums document each value in "Parameters:" section
+- [ ] Writing is concise and action-oriented
+- [ ] No documentation added to private methods (starting with `_`)
+- [ ] Existing complete docstrings were left unchanged

From 7e6e3031e721f04dedb2700aed4cf7b29d88e8ea Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Aleix=20Conchillo=20Flaqu=C3=A9?= <aleix@daily.co>
Date: Thu, 22 Jan 2026 13:29:12 -0800
Subject: [PATCH 49/51] claude: add pr-description skill

---
 .claude/skills/pr-description/SKILL.md | 128 +++++++++++++++++++++++++
 1 file changed, 128 insertions(+)
 create mode 100644 .claude/skills/pr-description/SKILL.md

diff --git a/.claude/skills/pr-description/SKILL.md b/.claude/skills/pr-description/SKILL.md
new file mode 100644
index 000000000..666cf2bd1
--- /dev/null
+++ b/.claude/skills/pr-description/SKILL.md
@@ -0,0 +1,128 @@
+---
+name: pr-description
+description: Update a GitHub PR description with a summary of changes
+---
+
+Update a GitHub pull request description based on the changes in the PR.
+
+## Arguments
+
+```
+/pr-description <PR_NUMBER> [--fixes <ISSUE_NUMBERS>]
+```
+
+- `PR_NUMBER` (required): The pull request number to update
+- `--fixes` (optional): Comma-separated issue numbers that this PR fixes (e.g., `--fixes 123,456`)
+
+Examples:
+- `/pr-description 3534`
+- `/pr-description 3534 --fixes 123`
+- `/pr-description 3534 --fixes 123,456,789`
+
+## Instructions
+
+1. First, gather information about the PR:
+   - Use GitHub plugin to get PR details (title, current description, base branch)
+   - Use local git to get commits: `git log main..HEAD --oneline`
+   - Use local git to get the diff: `git diff main..HEAD`
+   - Parse any `--fixes` argument for issue numbers
+
+2. Check the existing PR description:
+   - If it already has a complete, accurate description that reflects the changes, do nothing
+   - If it's missing sections, incomplete, or outdated compared to the actual changes, proceed to update
+   - If it only has the template placeholder text, generate a full description
+
+3. Analyze the changes:
+   - Understand the purpose of each commit
+   - Identify any breaking changes (API changes, removed features, behavior changes)
+   - Look for new features, bug fixes, refactoring, or documentation changes
+   - Collect issue numbers from:
+     - The `--fixes` argument (if provided)
+     - Commit messages (patterns like "Fixes #123", "Closes #456", "Resolves #789")
+
+4. Generate or update the PR description with these sections:
+
+## PR Description Format
+
+### Summary (always include)
+
+Brief bullet points describing what changed and why. Focus on the *purpose* and *impact*, not implementation details.
+
+```markdown
+## Summary
+
+- Added X to enable Y
+- Fixed bug where Z would happen
+- Refactored W for better maintainability
+```
+
+### Breaking Changes (include only if applicable)
+
+Document any changes that affect existing users or APIs.
+
+```markdown
+## Breaking Changes
+
+- `ClassName.method()` now requires a `param` argument
+- Removed deprecated `old_function()` - use `new_function()` instead
+```
+
+### Testing (include when non-obvious)
+
+How to verify the changes work. Skip for trivial changes.
+
+```markdown
+## Testing
+
+- Run `uv run pytest tests/test_feature.py` to verify the fix
+- Example usage: `uv run examples/new_feature.py`
+```
+
+### Fixes (include if issues are provided or found in commits)
+
+List issues this PR fixes. GitHub will automatically close these issues when the PR is merged.
+
+```markdown
+## Fixes
+
+- Fixes #123
+- Fixes #456
+```
+
+Note: Use "Fixes #X" format (not "Closes" or "Resolves") for consistency. Each issue should be on its own line with "Fixes" to ensure GitHub auto-closes them.
+
+## Guidelines
+
+- **Be concise** - Reviewers should understand the PR in 30 seconds
+- **Focus on why** - The diff shows *what* changed, explain *why*
+- **Skip empty sections** - Only include sections that have content
+- **Use bullet points** - Easier to scan than paragraphs
+- **Don't duplicate the diff** - Avoid listing every file or line changed
+
+## Example Output
+
+```markdown
+## Summary
+
+- Added `/docstring` skill for documenting Python modules with Google-style docstrings
+- Skill finds classes by name and handles conflicts when multiple matches exist
+- Skips already-documented code to avoid unnecessary changes
+
+## Testing
+
+/docstring ClassName
+
+## Fixes
+
+- Fixes #123
+```
+
+## Checklist
+
+Before updating the PR:
+
+- [ ] Verified existing description needs updating (not already complete)
+- [ ] Summary accurately reflects the changes
+- [ ] Breaking changes are clearly documented (if any)
+- [ ] No unnecessary sections included
+- [ ] Description is concise and scannable

From bcb019e8abd2abc5bd8c8b26459e3bba38ed2435 Mon Sep 17 00:00:00 2001
From: Mark Backman <mark@daily.co>
Date: Fri, 23 Jan 2026 18:47:34 -0500
Subject: [PATCH 50/51] Add TTFB metrics for STT services (#3495)

---
 changelog/3495.changed.2.md                   |   1 +
 changelog/3495.changed.md                     |   1 +
 src/pipecat/frames/frames.py                  |   3 +
 src/pipecat/services/assemblyai/stt.py        |   3 +-
 src/pipecat/services/aws/stt.py               |   2 -
 src/pipecat/services/azure/stt.py             |   2 -
 src/pipecat/services/cartesia/stt.py          |   9 +-
 src/pipecat/services/deepgram/flux/stt.py     |   3 +
 src/pipecat/services/deepgram/stt.py          |  17 +-
 .../services/deepgram/stt_sagemaker.py        |  10 +-
 src/pipecat/services/elevenlabs/stt.py        |  15 +-
 src/pipecat/services/fal/stt.py               |   2 -
 src/pipecat/services/gladia/stt.py            |   2 -
 src/pipecat/services/google/stt.py            |   2 -
 src/pipecat/services/gradium/stt.py           |   1 -
 src/pipecat/services/hathora/stt.py           |   2 -
 src/pipecat/services/nvidia/stt.py            |   4 -
 src/pipecat/services/sarvam/stt.py            |  44 +++-
 src/pipecat/services/soniox/stt.py            |   9 +-
 src/pipecat/services/stt_service.py           | 234 ++++++++++++++++++
 src/pipecat/services/whisper/base_stt.py      |   2 -
 src/pipecat/services/whisper/stt.py           |   4 -
 22 files changed, 309 insertions(+), 63 deletions(-)
 create mode 100644 changelog/3495.changed.2.md
 create mode 100644 changelog/3495.changed.md

diff --git a/changelog/3495.changed.2.md b/changelog/3495.changed.2.md
new file mode 100644
index 000000000..cf1f526b8
--- /dev/null
+++ b/changelog/3495.changed.2.md
@@ -0,0 +1 @@
+- `SarvamSTTService` now defaults `vad_signals` and `high_vad_sensitivity` to `None` (omitted from connection parameters), improving latency by ~300ms compared to the previous defaults.
diff --git a/changelog/3495.changed.md b/changelog/3495.changed.md
new file mode 100644
index 000000000..c690ebc09
--- /dev/null
+++ b/changelog/3495.changed.md
@@ -0,0 +1 @@
+- Improved the STT TTFB (Time To First Byte) measurement, reporting the delay between when the user stops speaking and when the final transcription is received. Note: Unlike traditional TTFB which measures from a discrete request, STT services receive continuous audio input—so we measure from speech end to final transcript, which captures the latency that matters for voice AI applications. In support of this change, added `finalized` field to `TranscriptionFrame` to indicate when a transcript is the final result for an utterance.
diff --git a/src/pipecat/frames/frames.py b/src/pipecat/frames/frames.py
index a27bb780e..26f9af9d2 100644
--- a/src/pipecat/frames/frames.py
+++ b/src/pipecat/frames/frames.py
@@ -426,12 +426,15 @@ class TranscriptionFrame(TextFrame):
         timestamp: When the transcription occurred.
         language: Detected or specified language of the speech.
         result: Raw result from the STT service.
+        finalized: Whether this is the final transcription for an utterance.
+            Set by STT services that support commit/finalize signals.
     """
 
     user_id: str
     timestamp: str
     language: Optional[Language] = None
     result: Optional[Any] = None
+    finalized: bool = False
 
     def __str__(self):
         return f"{self.name}(user: {self.user_id}, text: [{self.text}], language: {self.language}, timestamp: {self.timestamp})"
diff --git a/src/pipecat/services/assemblyai/stt.py b/src/pipecat/services/assemblyai/stt.py
index 3fde9491c..99b4f423d 100644
--- a/src/pipecat/services/assemblyai/stt.py
+++ b/src/pipecat/services/assemblyai/stt.py
@@ -161,7 +161,7 @@ class AssemblyAISTTService(WebsocketSTTService):
         """
         await super().process_frame(frame, direction)
         if isinstance(frame, VADUserStartedSpeakingFrame):
-            await self.start_ttfb_metrics()
+            pass
         elif isinstance(frame, VADUserStoppedSpeakingFrame):
             if (
                 self._vad_force_turn_endpoint
@@ -354,7 +354,6 @@ class AssemblyAISTTService(WebsocketSTTService):
         """Handle transcription results."""
         if not message.transcript:
             return
-        await self.stop_ttfb_metrics()
         if message.end_of_turn and (
             not self._connection_params.formatted_finals or message.turn_is_formatted
         ):
diff --git a/src/pipecat/services/aws/stt.py b/src/pipecat/services/aws/stt.py
index 5ae99b02f..8beefe051 100644
--- a/src/pipecat/services/aws/stt.py
+++ b/src/pipecat/services/aws/stt.py
@@ -158,7 +158,6 @@ class AWSTranscribeSTTService(WebsocketSTTService):
                 await self._websocket.send(event_message)
                 # Start metrics after first chunk sent
                 await self.start_processing_metrics()
-                await self.start_ttfb_metrics()
             except Exception as e:
                 yield ErrorFrame(error=f"Error sending audio: {e}")
 
@@ -470,7 +469,6 @@ class AWSTranscribeSTTService(WebsocketSTTService):
                             is_final = not result.get("IsPartial", True)
 
                             if transcript:
-                                await self.stop_ttfb_metrics()
                                 if is_final:
                                     await self.push_frame(
                                         TranscriptionFrame(
diff --git a/src/pipecat/services/azure/stt.py b/src/pipecat/services/azure/stt.py
index aaa1e2da4..d45eb71df 100644
--- a/src/pipecat/services/azure/stt.py
+++ b/src/pipecat/services/azure/stt.py
@@ -116,7 +116,6 @@ class AzureSTTService(STTService):
         """
         try:
             await self.start_processing_metrics()
-            await self.start_ttfb_metrics()
             if self._audio_stream:
                 self._audio_stream.write(audio)
             yield None
@@ -191,7 +190,6 @@ class AzureSTTService(STTService):
         self, transcript: str, is_final: bool, language: Optional[Language] = None
     ):
         """Handle a transcription result with tracing."""
-        await self.stop_ttfb_metrics()
         await self.stop_processing_metrics()
 
     def _on_handle_recognized(self, event):
diff --git a/src/pipecat/services/cartesia/stt.py b/src/pipecat/services/cartesia/stt.py
index 625df6366..2fb08b981 100644
--- a/src/pipecat/services/cartesia/stt.py
+++ b/src/pipecat/services/cartesia/stt.py
@@ -207,9 +207,8 @@ class CartesiaSTTService(WebsocketSTTService):
         await super().cancel(frame)
         await self._disconnect()
 
-    async def start_metrics(self):
+    async def _start_metrics(self):
         """Start performance metrics collection for transcription processing."""
-        await self.start_ttfb_metrics()
         await self.start_processing_metrics()
 
     async def process_frame(self, frame: Frame, direction: FrameDirection):
@@ -222,10 +221,13 @@ class CartesiaSTTService(WebsocketSTTService):
         await super().process_frame(frame, direction)
 
         if isinstance(frame, VADUserStartedSpeakingFrame):
-            await self.start_metrics()
+            # Reset finalize state for new utterance
+            self.set_finalize_pending(False)
+            await self._start_metrics()
         elif isinstance(frame, VADUserStoppedSpeakingFrame):
             # Send finalize command to flush the transcription session
             if self._websocket and self._websocket.state is State.OPEN:
+                self.set_finalize_pending(True)
                 await self._websocket.send("finalize")
 
     async def run_stt(self, audio: bytes) -> AsyncGenerator[Frame, None]:
@@ -342,7 +344,6 @@ class CartesiaSTTService(WebsocketSTTService):
                 pass
 
         if len(transcript) > 0:
-            await self.stop_ttfb_metrics()
             if is_final:
                 await self.push_frame(
                     TranscriptionFrame(
diff --git a/src/pipecat/services/deepgram/flux/stt.py b/src/pipecat/services/deepgram/flux/stt.py
index 045af9811..756827171 100644
--- a/src/pipecat/services/deepgram/flux/stt.py
+++ b/src/pipecat/services/deepgram/flux/stt.py
@@ -659,6 +659,8 @@ class DeepgramFluxSTTService(WebsocketSTTService):
         average_confidence = self._calculate_average_confidence(data)
 
         if not self._params.min_confidence or average_confidence > self._params.min_confidence:
+            # EndOfTurn means Flux has determined the turn is complete,
+            # so this TranscriptionFrame is always finalized
             await self.push_frame(
                 TranscriptionFrame(
                     transcript,
@@ -666,6 +668,7 @@ class DeepgramFluxSTTService(WebsocketSTTService):
                     time_now_iso8601(),
                     self._language,
                     result=data,
+                    finalized=True,
                 )
             )
         else:
diff --git a/src/pipecat/services/deepgram/stt.py b/src/pipecat/services/deepgram/stt.py
index 2469b25a3..20dfba724 100644
--- a/src/pipecat/services/deepgram/stt.py
+++ b/src/pipecat/services/deepgram/stt.py
@@ -276,9 +276,8 @@ class DeepgramSTTService(STTService):
             # GH issue: https://github.com/deepgram/deepgram-python-sdk/issues/570
             await self._connection.finish()
 
-    async def start_metrics(self):
-        """Start TTFB and processing metrics collection."""
-        await self.start_ttfb_metrics()
+    async def _start_metrics(self):
+        """Start processing metrics collection for this utterance."""
         await self.start_processing_metrics()
 
     async def _on_error(self, *args, **kwargs):
@@ -292,7 +291,7 @@ class DeepgramSTTService(STTService):
         await self._connect()
 
     async def _on_speech_started(self, *args, **kwargs):
-        await self.start_metrics()
+        await self._start_metrics()
         await self._call_event_handler("on_speech_started", *args, **kwargs)
         await self.broadcast_frame(UserStartedSpeakingFrame)
         if self._should_interrupt:
@@ -320,8 +319,12 @@ class DeepgramSTTService(STTService):
             language = result.channel.alternatives[0].languages[0]
             language = Language(language)
         if len(transcript) > 0:
-            await self.stop_ttfb_metrics()
             if is_final:
+                # Check if this response is from a finalize() call.
+                # Only mark as finalized when both we requested it AND Deepgram confirms it.
+                from_finalize = getattr(result, "from_finalize", False)
+                if from_finalize:
+                    self.confirm_finalize()
                 await self.push_frame(
                     TranscriptionFrame(
                         transcript,
@@ -356,8 +359,10 @@ class DeepgramSTTService(STTService):
 
         if isinstance(frame, VADUserStartedSpeakingFrame) and not self.vad_enabled:
             # Start metrics if Deepgram VAD is disabled & pipeline VAD has detected speech
-            await self.start_metrics()
+            await self._start_metrics()
         elif isinstance(frame, VADUserStoppedSpeakingFrame):
             # https://developers.deepgram.com/docs/finalize
+            # Mark that we're awaiting a from_finalize response
+            self.request_finalize()
             await self._connection.finalize()
             logger.trace(f"Triggered finalize event on: {frame.name=}, {direction=}")
diff --git a/src/pipecat/services/deepgram/stt_sagemaker.py b/src/pipecat/services/deepgram/stt_sagemaker.py
index 7eb6a072b..0f7e6f988 100644
--- a/src/pipecat/services/deepgram/stt_sagemaker.py
+++ b/src/pipecat/services/deepgram/stt_sagemaker.py
@@ -363,9 +363,6 @@ class DeepgramSageMakerSTTService(STTService):
         if not transcript.strip():
             return
 
-        # Stop TTFB metrics on first transcript
-        await self.stop_ttfb_metrics()
-
         is_final = parsed.get("is_final", False)
         speech_final = parsed.get("speech_final", False)
 
@@ -417,9 +414,8 @@ class DeepgramSageMakerSTTService(STTService):
         """
         pass
 
-    async def start_metrics(self):
-        """Start TTFB and processing metrics collection."""
-        await self.start_ttfb_metrics()
+    async def _start_metrics(self):
+        """Start processing metrics collection."""
         await self.start_processing_metrics()
 
     async def process_frame(self, frame: Frame, direction: FrameDirection):
@@ -433,7 +429,7 @@ class DeepgramSageMakerSTTService(STTService):
 
         # Start metrics when user starts speaking (if VAD is not provided by Deepgram)
         if isinstance(frame, VADUserStartedSpeakingFrame):
-            await self.start_metrics()
+            await self._start_metrics()
         elif isinstance(frame, VADUserStoppedSpeakingFrame):
             # Send finalize message to Deepgram when user stops speaking
             # This tells Deepgram to flush any remaining audio and return final results
diff --git a/src/pipecat/services/elevenlabs/stt.py b/src/pipecat/services/elevenlabs/stt.py
index 8f9020aa7..a1be56a0d 100644
--- a/src/pipecat/services/elevenlabs/stt.py
+++ b/src/pipecat/services/elevenlabs/stt.py
@@ -310,7 +310,6 @@ class ElevenLabsSTTService(SegmentedSTTService):
         self, transcript: str, is_final: bool, language: Optional[str] = None
     ):
         """Handle a transcription result with tracing."""
-        await self.stop_ttfb_metrics()
         await self.stop_processing_metrics()
 
     async def run_stt(self, audio: bytes) -> AsyncGenerator[Frame, None]:
@@ -328,7 +327,6 @@ class ElevenLabsSTTService(SegmentedSTTService):
         """
         try:
             await self.start_processing_metrics()
-            await self.start_ttfb_metrics()
 
             # Upload audio and get transcription result directly
             result = await self._transcribe_audio(audio)
@@ -539,9 +537,8 @@ class ElevenLabsRealtimeSTTService(WebsocketSTTService):
         await super().cancel(frame)
         await self._disconnect()
 
-    async def start_metrics(self):
+    async def _start_metrics(self):
         """Start performance metrics collection for transcription processing."""
-        await self.start_ttfb_metrics()
         await self.start_processing_metrics()
 
     async def process_frame(self, frame: Frame, direction: FrameDirection):
@@ -554,13 +551,17 @@ class ElevenLabsRealtimeSTTService(WebsocketSTTService):
         await super().process_frame(frame, direction)
 
         if isinstance(frame, VADUserStartedSpeakingFrame):
+            # Reset finalize state for new utterance
+            self.set_finalize_pending(False)
             # Start metrics when user starts speaking
-            await self.start_metrics()
+            await self._start_metrics()
         elif isinstance(frame, VADUserStoppedSpeakingFrame):
             # Send commit when user stops speaking (manual commit mode)
             if self._params.commit_strategy == CommitStrategy.MANUAL:
                 if self._websocket and self._websocket.state is State.OPEN:
                     try:
+                        # Mark that the next committed transcript should be finalized
+                        self.set_finalize_pending(True)
                         commit_message = {
                             "message_type": "input_audio_chunk",
                             "audio_base_64": "",
@@ -764,8 +765,6 @@ class ElevenLabsRealtimeSTTService(WebsocketSTTService):
         if not text:
             return
 
-        await self.stop_ttfb_metrics()
-
         # Get language if provided
         language = data.get("language_code")
 
@@ -803,7 +802,6 @@ class ElevenLabsRealtimeSTTService(WebsocketSTTService):
         if not text:
             return
 
-        await self.stop_ttfb_metrics()
         await self.stop_processing_metrics()
 
         # Get language if provided
@@ -845,7 +843,6 @@ class ElevenLabsRealtimeSTTService(WebsocketSTTService):
         if not text:
             return
 
-        await self.stop_ttfb_metrics()
         await self.stop_processing_metrics()
 
         # Get language if provided
diff --git a/src/pipecat/services/fal/stt.py b/src/pipecat/services/fal/stt.py
index a71915743..114865778 100644
--- a/src/pipecat/services/fal/stt.py
+++ b/src/pipecat/services/fal/stt.py
@@ -249,7 +249,6 @@ class FalSTTService(SegmentedSTTService):
         self, transcript: str, is_final: bool, language: Optional[str] = None
     ):
         """Handle a transcription result with tracing."""
-        await self.stop_ttfb_metrics()
         await self.stop_processing_metrics()
 
     async def run_stt(self, audio: bytes) -> AsyncGenerator[Frame, None]:
@@ -267,7 +266,6 @@ class FalSTTService(SegmentedSTTService):
         """
         try:
             await self.start_processing_metrics()
-            await self.start_ttfb_metrics()
 
             # Send to Fal directly (audio is already in WAV format from base class)
             data_uri = fal_client.encode(audio, "audio/x-wav")
diff --git a/src/pipecat/services/gladia/stt.py b/src/pipecat/services/gladia/stt.py
index 8c5ec7aa4..e4229fe97 100644
--- a/src/pipecat/services/gladia/stt.py
+++ b/src/pipecat/services/gladia/stt.py
@@ -385,7 +385,6 @@ class GladiaSTTService(WebsocketSTTService):
         Yields:
             None (processing is handled asynchronously via WebSocket).
         """
-        await self.start_ttfb_metrics()
         await self.start_processing_metrics()
 
         # Add audio to buffer
@@ -513,7 +512,6 @@ class GladiaSTTService(WebsocketSTTService):
     async def _handle_transcription(
         self, transcript: str, is_final: bool, language: Optional[str] = None
     ):
-        await self.stop_ttfb_metrics()
         await self.stop_processing_metrics()
 
     async def _on_speech_started(self):
diff --git a/src/pipecat/services/google/stt.py b/src/pipecat/services/google/stt.py
index dcf0adf89..edb2ccec0 100644
--- a/src/pipecat/services/google/stt.py
+++ b/src/pipecat/services/google/stt.py
@@ -823,7 +823,6 @@ class GoogleSTTService(STTService):
         """
         if self._streaming_task:
             # Queue the audio data
-            await self.start_ttfb_metrics()
             await self.start_processing_metrics()
             await self._request_queue.put(audio)
         yield None
@@ -875,7 +874,6 @@ class GoogleSTTService(STTService):
                         )
                     else:
                         self._last_transcript_was_final = False
-                        await self.stop_ttfb_metrics()
                         await self.push_frame(
                             InterimTranscriptionFrame(
                                 transcript,
diff --git a/src/pipecat/services/gradium/stt.py b/src/pipecat/services/gradium/stt.py
index b66b18070..1c378ff9f 100644
--- a/src/pipecat/services/gradium/stt.py
+++ b/src/pipecat/services/gradium/stt.py
@@ -122,7 +122,6 @@ class GradiumSTTService(WebsocketSTTService):
             None (processing handled via WebSocket messages).
         """
         self._audio_buffer.extend(audio)
-        await self.start_ttfb_metrics()
         await self.start_processing_metrics()
 
         while len(self._audio_buffer) >= self._chunk_size_bytes:
diff --git a/src/pipecat/services/hathora/stt.py b/src/pipecat/services/hathora/stt.py
index 8aadb6b65..81e8f0ea6 100644
--- a/src/pipecat/services/hathora/stt.py
+++ b/src/pipecat/services/hathora/stt.py
@@ -111,7 +111,6 @@ class HathoraSTTService(SegmentedSTTService):
         """
         try:
             await self.start_processing_metrics()
-            await self.start_ttfb_metrics()
 
             url = f"{self._base_url}"
 
@@ -153,7 +152,6 @@ class HathoraSTTService(SegmentedSTTService):
                         result=response,
                     )
 
-            await self.stop_ttfb_metrics()
             await self.stop_processing_metrics()
 
         except Exception as e:
diff --git a/src/pipecat/services/nvidia/stt.py b/src/pipecat/services/nvidia/stt.py
index 7d52f5130..3af3fe78c 100644
--- a/src/pipecat/services/nvidia/stt.py
+++ b/src/pipecat/services/nvidia/stt.py
@@ -307,7 +307,6 @@ class NvidiaSTTService(STTService):
 
             transcript = result.alternatives[0].transcript
             if transcript and len(transcript) > 0:
-                await self.stop_ttfb_metrics()
                 if result.is_final:
                     await self.stop_processing_metrics()
                     await self.push_frame(
@@ -344,7 +343,6 @@ class NvidiaSTTService(STTService):
         Yields:
             None - transcription results are pushed to the pipeline via frames.
         """
-        await self.start_ttfb_metrics()
         await self.start_processing_metrics()
         await self._queue.put(audio)
         yield None
@@ -598,12 +596,10 @@ class NvidiaSegmentedSTTService(SegmentedSTTService):
             assert self._config is not None, "Recognition config not created"
 
             await self.start_processing_metrics()
-            await self.start_ttfb_metrics()
 
             # Process audio with NVIDIA Riva ASR - explicitly request non-future response
             raw_response = self._asr_service.offline_recognize(audio, self._config, future=False)
 
-            await self.stop_ttfb_metrics()
             await self.stop_processing_metrics()
 
             # Process the response - handle different possible return types
diff --git a/src/pipecat/services/sarvam/stt.py b/src/pipecat/services/sarvam/stt.py
index f4e5c676b..96447dce9 100644
--- a/src/pipecat/services/sarvam/stt.py
+++ b/src/pipecat/services/sarvam/stt.py
@@ -15,9 +15,15 @@ from pipecat.frames.frames import (
     CancelFrame,
     EndFrame,
     ErrorFrame,
+    Frame,
     StartFrame,
     TranscriptionFrame,
+    UserStartedSpeakingFrame,
+    UserStoppedSpeakingFrame,
+    VADUserStartedSpeakingFrame,
+    VADUserStoppedSpeakingFrame,
 )
+from pipecat.processors.frame_processor import FrameDirection
 from pipecat.services.sarvam._sdk import sdk_headers
 from pipecat.services.stt_service import STTService
 from pipecat.transcriptions.language import Language, resolve_language
@@ -75,14 +81,14 @@ class SarvamSTTService(STTService):
             language: Target language for transcription. Defaults to None (required for saarika models).
             prompt: Optional prompt to guide translation style/context for STT-Translate models.
                    Only applicable to saaras (STT-Translate) models. Defaults to None.
-            vad_signals: Enable VAD signals in response. Defaults to True.
-            high_vad_sensitivity: Enable high VAD (Voice Activity Detection) sensitivity. Defaults to False.
+            vad_signals: Enable VAD signals in response. Defaults to None.
+            high_vad_sensitivity: Enable high VAD (Voice Activity Detection) sensitivity. Defaults to None.
         """
 
         language: Optional[Language] = None
         prompt: Optional[str] = None
-        vad_signals: bool = True
-        high_vad_sensitivity: bool = False
+        vad_signals: bool = None
+        high_vad_sensitivity: bool = None
 
     def __init__(
         self,
@@ -155,6 +161,7 @@ class SarvamSTTService(STTService):
         self._websocket_context = None
         self._socket_client = None
         self._receive_task = None
+        logger.info(f"Sarvam STT initialized with SDK headers: {self._sdk_headers}")
 
     def language_to_service_language(self, language: Language) -> str:
         """Convert pipecat Language enum to Sarvam's language code.
@@ -175,6 +182,24 @@ class SarvamSTTService(STTService):
         """
         return True
 
+    async def process_frame(self, frame: Frame, direction: FrameDirection):
+        """Process incoming frames.
+
+        Handles VAD frames for TTFB tracking when using Pipecat's VAD
+        instead of Sarvam's built-in VAD.
+        """
+        await super().process_frame(frame, direction)
+
+        # Only handle VAD frames when not using Sarvam's VAD signals
+        if not self._vad_signals:
+            if isinstance(frame, VADUserStartedSpeakingFrame):
+                self.set_finalize_pending(False)
+                await self._start_metrics()
+            elif isinstance(frame, VADUserStoppedSpeakingFrame):
+                if self._socket_client:
+                    self.set_finalize_pending(True)
+                    await self._socket_client.flush()
+
     async def set_language(self, language: Language):
         """Set the recognition language and reconnect.
 
@@ -411,16 +436,18 @@ class SarvamSTTService(STTService):
                 logger.debug(f"VAD Signal: {signal}, Occurred at: {timestamp}")
 
                 if signal == "START_SPEECH":
-                    await self.start_metrics()
+                    await self._start_metrics()
                     logger.debug("User started speaking")
                     await self._call_event_handler("on_speech_started")
+                    await self.broadcast_frame(UserStartedSpeakingFrame)
+                    await self.push_interruption_task_frame_and_wait()
 
                 elif signal == "END_SPEECH":
                     logger.debug("User stopped speaking")
                     await self._call_event_handler("on_speech_stopped")
+                    await self.broadcast_frame(UserStoppedSpeakingFrame)
 
             elif message.type == "data":
-                await self.stop_ttfb_metrics()
                 transcript = message.data.transcript
                 language_code = message.data.language_code
                 # Prefer language from message (auto-detected for translate models). Fallback to configured.
@@ -482,7 +509,6 @@ class SarvamSTTService(STTService):
         }
         return mapping.get(language_code, Language.HI_IN)
 
-    async def start_metrics(self):
-        """Start TTFB and processing metrics collection."""
-        await self.start_ttfb_metrics()
+    async def _start_metrics(self):
+        """Start processing metrics collection."""
         await self.start_processing_metrics()
diff --git a/src/pipecat/services/soniox/stt.py b/src/pipecat/services/soniox/stt.py
index 476ae0762..6f8c92d26 100644
--- a/src/pipecat/services/soniox/stt.py
+++ b/src/pipecat/services/soniox/stt.py
@@ -21,7 +21,7 @@ from pipecat.frames.frames import (
     InterimTranscriptionFrame,
     StartFrame,
     TranscriptionFrame,
-    UserStoppedSpeakingFrame,
+    VADUserStoppedSpeakingFrame,
 )
 from pipecat.processors.frame_processor import FrameDirection
 from pipecat.services.stt_service import WebsocketSTTService
@@ -162,7 +162,7 @@ class SonioxSTTService(WebsocketSTTService):
             sample_rate: Audio sample rate.
             params: Additional configuration parameters, such as language hints, context and
                 speaker diarization.
-            vad_force_turn_endpoint: Listen to `UserStoppedSpeakingFrame` to send finalize message to Soniox. If disabled, Soniox will detect the end of the speech.
+            vad_force_turn_endpoint: Listen to `VADUserStoppedSpeakingFrame` to send finalize message to Soniox. If disabled, Soniox will detect the end of the speech.
             **kwargs: Additional arguments passed to the STTService.
         """
         super().__init__(sample_rate=sample_rate, **kwargs)
@@ -247,7 +247,7 @@ class SonioxSTTService(WebsocketSTTService):
         """
         await super().process_frame(frame, direction)
 
-        if isinstance(frame, UserStoppedSpeakingFrame) and self._vad_force_turn_endpoint:
+        if isinstance(frame, VADUserStoppedSpeakingFrame) and self._vad_force_turn_endpoint:
             # Send finalize message to Soniox so we get the final tokens asap.
             if self._websocket and self._websocket.state is State.OPEN:
                 await self._websocket.send(FINALIZE_MESSAGE)
@@ -374,12 +374,15 @@ class SonioxSTTService(WebsocketSTTService):
         async def send_endpoint_transcript():
             if self._final_transcription_buffer:
                 text = "".join(map(lambda token: token["text"], self._final_transcription_buffer))
+                # Soniox only pushes TranscriptionFrame when an end token is received,
+                # so every TranscriptionFrame is inherently finalized
                 await self.push_frame(
                     TranscriptionFrame(
                         text=text,
                         user_id=self._user_id,
                         timestamp=time_now_iso8601(),
                         result=self._final_transcription_buffer,
+                        finalized=True,
                     )
                 )
                 await self._handle_transcription(text, is_final=True)
diff --git a/src/pipecat/services/stt_service.py b/src/pipecat/services/stt_service.py
index 46c40ae70..24b1aafcf 100644
--- a/src/pipecat/services/stt_service.py
+++ b/src/pipecat/services/stt_service.py
@@ -6,7 +6,9 @@
 
 """Base classes for Speech-to-Text services with continuous and segmented processing."""
 
+import asyncio
 import io
+import time
 import wave
 from abc import abstractmethod
 from typing import Any, AsyncGenerator, Dict, Mapping, Optional
@@ -17,12 +19,17 @@ from pipecat.frames.frames import (
     AudioRawFrame,
     ErrorFrame,
     Frame,
+    InterruptionFrame,
+    MetricsFrame,
+    SpeechControlParamsFrame,
     StartFrame,
     STTMuteFrame,
     STTUpdateSettingsFrame,
+    TranscriptionFrame,
     VADUserStartedSpeakingFrame,
     VADUserStoppedSpeakingFrame,
 )
+from pipecat.metrics.metrics import TTFBMetricsData
 from pipecat.processors.frame_processor import FrameDirection
 from pipecat.services.ai_service import AIService
 from pipecat.services.websocket_service import WebsocketService
@@ -61,6 +68,8 @@ class STTService(AIService):
         audio_passthrough=True,
         # STT input sample rate
         sample_rate: Optional[int] = None,
+        # STT TTFB timeout - time to wait after VAD stop before reporting TTFB
+        stt_ttfb_timeout: float = 2.0,
         **kwargs,
     ):
         """Initialize the STT service.
@@ -70,6 +79,12 @@ class STTService(AIService):
                 Defaults to True.
             sample_rate: The sample rate for audio input. If None, will be determined
                 from the start frame.
+            stt_ttfb_timeout: Time in seconds to wait after VAD stop before reporting
+                TTFB. This delay allows the final transcription to arrive. Defaults to 2.0.
+                Note: STT "TTFB" differs from traditional TTFB (which measures from a discrete
+                request to first response byte). Since STT receives continuous audio, we measure
+                from when the user stops speaking to when the final transcript arrives—capturing
+                the latency that matters for voice AI applications.
             **kwargs: Additional arguments passed to the parent AIService.
         """
         super().__init__(**kwargs)
@@ -81,6 +96,16 @@ class STTService(AIService):
         self._muted: bool = False
         self._user_id: str = ""
 
+        # STT TTFB tracking state
+        self._stt_ttfb_timeout = stt_ttfb_timeout
+        self._ttfb_timeout_task: Optional[asyncio.Task] = None
+        self._vad_stop_secs: Optional[float] = None
+        self._speech_end_time: Optional[float] = None
+        self._user_speaking: bool = False
+        self._last_transcription_time: Optional[float] = None
+        self._finalize_pending: bool = False
+        self._finalize_requested: bool = False
+
         self._register_event_handler("on_connected")
         self._register_event_handler("on_disconnected")
         self._register_event_handler("on_connection_error")
@@ -94,6 +119,44 @@ class STTService(AIService):
         """
         return self._muted
 
+    def set_finalize_pending(self, value: bool):
+        """Set whether the next TranscriptionFrame should be marked as finalized.
+
+        When True, the next TranscriptionFrame pushed will have its `finalized`
+        field set to True, and this flag will automatically reset to False.
+        This is used to signal that a transcript is the final result for an
+        utterance, enabling immediate TTFB reporting.
+
+        Args:
+            value: True to mark the next transcription as finalized.
+        """
+        self._finalize_pending = value
+
+    def request_finalize(self):
+        """Mark that a finalize request has been sent, awaiting server confirmation.
+
+        For providers that require server confirmation before marking transcripts
+        as finalized (e.g., Deepgram's from_finalize field), call this when sending
+        the finalize request. Then call confirm_finalize() when the server confirms.
+
+        This is an alternative to set_finalize_pending() for providers that need
+        two-step finalization.
+        """
+        self._finalize_requested = True
+
+    def confirm_finalize(self):
+        """Confirm that the server has acknowledged the finalize request.
+
+        Call this when the server response confirms finalization (e.g., Deepgram's
+        from_finalize=True). The next TranscriptionFrame pushed will be marked
+        as finalized.
+
+        Only has effect if request_finalize() was previously called.
+        """
+        if self._finalize_requested:
+            self._finalize_pending = True
+            self._finalize_requested = False
+
     @property
     def sample_rate(self) -> int:
         """Get the current sample rate for audio processing.
@@ -144,6 +207,11 @@ class STTService(AIService):
         self._sample_rate = self._init_sample_rate or frame.audio_in_sample_rate
         self._tracing_enabled = frame.enable_tracing
 
+    async def cleanup(self):
+        """Clean up STT service resources."""
+        await super().cleanup()
+        await self._cancel_ttfb_timeout()
+
     async def _update_settings(self, settings: Mapping[str, Any]):
         logger.info(f"Updating STT settings: {self._settings}")
         for key, value in settings.items():
@@ -206,14 +274,166 @@ class STTService(AIService):
             await self.process_audio_frame(frame, direction)
             if self._audio_passthrough:
                 await self.push_frame(frame, direction)
+        elif isinstance(frame, SpeechControlParamsFrame):
+            await self._handle_speech_control_params(frame)
+            await self.push_frame(frame, direction)
+        elif isinstance(frame, VADUserStartedSpeakingFrame):
+            await self._handle_vad_user_started_speaking(frame)
+            await self.push_frame(frame, direction)
+        elif isinstance(frame, VADUserStoppedSpeakingFrame):
+            await self._handle_vad_user_stopped_speaking(frame)
+            await self.push_frame(frame, direction)
         elif isinstance(frame, STTUpdateSettingsFrame):
             await self._update_settings(frame.settings)
         elif isinstance(frame, STTMuteFrame):
             self._muted = frame.mute
             logger.debug(f"STT service {'muted' if frame.mute else 'unmuted'}")
+        elif isinstance(frame, InterruptionFrame):
+            await self._reset_stt_ttfb_state()
+            await self.push_frame(frame, direction)
         else:
             await self.push_frame(frame, direction)
 
+    async def push_frame(self, frame: Frame, direction: FrameDirection = FrameDirection.DOWNSTREAM):
+        """Push a frame downstream, tracking TranscriptionFrame timestamps for TTFB.
+
+        Stores the timestamp of each TranscriptionFrame for TTFB calculation.
+        If the frame is marked as finalized (either directly or via set_finalize_pending),
+        reports TTFB immediately and cancels any pending timeout. Otherwise, TTFB is
+        reported after a timeout.
+
+        Args:
+            frame: The frame to push.
+            direction: The direction to push the frame.
+        """
+        if isinstance(frame, TranscriptionFrame):
+            # Store the transcription time for TTFB calculation
+            self._last_transcription_time = time.time()
+
+            # Set finalized from pending state and auto-reset
+            if self._finalize_pending:
+                frame.finalized = True
+                self._finalize_pending = False
+
+            # If this is a finalized transcription, report TTFB immediately
+            if frame.finalized and self._speech_end_time is not None:
+                ttfb = self._last_transcription_time - self._speech_end_time
+                await self._emit_stt_ttfb_metric(ttfb)
+                # Cancel the timeout since we've already reported
+                await self._cancel_ttfb_timeout()
+                # Clear state
+                self._speech_end_time = None
+                self._last_transcription_time = None
+
+        await super().push_frame(frame, direction)
+
+    async def _handle_speech_control_params(self, frame: SpeechControlParamsFrame):
+        """Handle speech control parameters frame to extract VAD stop_secs.
+
+        Args:
+            frame: The speech control parameters frame.
+        """
+        if frame.vad_params is not None:
+            self._vad_stop_secs = frame.vad_params.stop_secs
+
+    async def _cancel_ttfb_timeout(self):
+        """Cancel any pending TTFB timeout task."""
+        if self._ttfb_timeout_task:
+            await self.cancel_task(self._ttfb_timeout_task)
+            self._ttfb_timeout_task = None
+
+    async def _reset_stt_ttfb_state(self):
+        """Reset STT TTFB measurement state.
+
+        Called when starting a new utterance or on interruption to ensure
+        we don't use stale state for TTFB calculations. This specifically guards
+        against the case where a TranscriptionFrame is received without corresponding
+        VADUserStartedSpeakingFrame and VADUserStoppedSpeakingFrame frames.
+
+        Note: Does not reset _user_speaking since InterruptionFrame can arrive
+        while user is still speaking.
+        """
+        await self._cancel_ttfb_timeout()
+        self._speech_end_time = None
+        self._last_transcription_time = None
+
+    async def _handle_vad_user_started_speaking(self, frame: VADUserStartedSpeakingFrame):
+        """Handle VAD user started speaking frame to start tracking transcriptions.
+
+        Cancels any pending TTFB timeout, resets TTFB tracking state, and marks user as speaking.
+
+        Args:
+            frame: The VAD user started speaking frame.
+        """
+        await self._reset_stt_ttfb_state()
+        self._user_speaking = True
+        self._finalize_requested = False
+
+    async def _handle_vad_user_stopped_speaking(self, frame: VADUserStoppedSpeakingFrame):
+        """Handle VAD user stopped speaking frame.
+
+        Calculates the actual speech end time and starts a timeout task to wait
+        for the final transcription before reporting TTFB.
+
+        Args:
+            frame: The VAD user stopped speaking frame.
+        """
+        self._user_speaking = False
+
+        # Skip TTFB measurement if we don't have VAD params
+        if self._vad_stop_secs is None:
+            return
+
+        # Calculate the actual speech end time (current time minus VAD stop delay).
+        # This approximates when the last user audio was sent to the STT service,
+        # which we use to measure against the eventual transcription response.
+        self._speech_end_time = time.time() - self._vad_stop_secs
+
+        # Start timeout task (any previous timeout was cancelled by VADUserStartedSpeakingFrame
+        # or InterruptionFrame)
+        self._ttfb_timeout_task = self.create_task(
+            self._ttfb_timeout_handler(), name="stt_ttfb_timeout"
+        )
+
+    async def _ttfb_timeout_handler(self):
+        """Wait for timeout then report TTFB using the last transcription timestamp.
+
+        This timeout allows the final transcription to arrive before we calculate
+        and report TTFB. If no transcription arrived, no TTFB is reported.
+        """
+        try:
+            await asyncio.sleep(self._stt_ttfb_timeout)
+
+            # Report TTFB if we have both speech end time and transcription time
+            if self._speech_end_time is not None and self._last_transcription_time is not None:
+                ttfb = self._last_transcription_time - self._speech_end_time
+                await self._emit_stt_ttfb_metric(ttfb)
+
+            # Clear state after reporting
+            self._speech_end_time = None
+            self._last_transcription_time = None
+        except asyncio.CancelledError:
+            # Task was cancelled (new utterance or interruption), which is expected behavior
+            pass
+        finally:
+            self._ttfb_timeout_task = None
+
+    async def _emit_stt_ttfb_metric(self, ttfb: float):
+        """Emit STT TTFB metric if value is non-negative.
+
+        Args:
+            ttfb: The TTFB value in seconds.
+        """
+        if ttfb >= 0:
+            logger.debug(f"{self} TTFB: {ttfb:.3f}s")
+            if self.metrics_enabled:
+                ttfb_data = TTFBMetricsData(
+                    processor=self.name,
+                    model=self.model_name,
+                    value=ttfb,
+                )
+                await super().push_frame(MetricsFrame(data=[ttfb_data]))
+
 
 class SegmentedSTTService(STTService):
     """STT service that processes speech in segments using VAD events.
@@ -250,6 +470,20 @@ class SegmentedSTTService(STTService):
         await super().start(frame)
         self._audio_buffer_size_1s = self.sample_rate * 2
 
+    async def push_frame(self, frame: Frame, direction: FrameDirection = FrameDirection.DOWNSTREAM):
+        """Push a frame, marking TranscriptionFrames as finalized.
+
+        Segmented STT services process complete speech segments and return a single
+        TranscriptionFrame per segment, so every transcription is inherently finalized.
+
+        Args:
+            frame: The frame to push.
+            direction: The direction of frame flow in the pipeline.
+        """
+        if isinstance(frame, TranscriptionFrame):
+            frame.finalized = True
+        await super().push_frame(frame, direction)
+
     async def process_frame(self, frame: Frame, direction: FrameDirection):
         """Process frames, handling VAD events and audio segmentation."""
         await super().process_frame(frame, direction)
diff --git a/src/pipecat/services/whisper/base_stt.py b/src/pipecat/services/whisper/base_stt.py
index 32e08de8e..79d2723f1 100644
--- a/src/pipecat/services/whisper/base_stt.py
+++ b/src/pipecat/services/whisper/base_stt.py
@@ -204,11 +204,9 @@ class BaseWhisperSTTService(SegmentedSTTService):
         """
         try:
             await self.start_processing_metrics()
-            await self.start_ttfb_metrics()
 
             response = await self._transcribe(audio)
 
-            await self.stop_ttfb_metrics()
             await self.stop_processing_metrics()
 
             text = response.text.strip()
diff --git a/src/pipecat/services/whisper/stt.py b/src/pipecat/services/whisper/stt.py
index 3865578e1..f11978cc2 100644
--- a/src/pipecat/services/whisper/stt.py
+++ b/src/pipecat/services/whisper/stt.py
@@ -289,7 +289,6 @@ class WhisperSTTService(SegmentedSTTService):
             return
 
         await self.start_processing_metrics()
-        await self.start_ttfb_metrics()
 
         # Divide by 32768 because we have signed 16-bit data.
         audio_float = np.frombuffer(audio, dtype=np.int16).astype(np.float32) / 32768.0
@@ -303,7 +302,6 @@ class WhisperSTTService(SegmentedSTTService):
             if segment.no_speech_prob < self._no_speech_prob:
                 text += f"{segment.text} "
 
-        await self.stop_ttfb_metrics()
         await self.stop_processing_metrics()
 
         if text:
@@ -388,7 +386,6 @@ class WhisperSTTServiceMLX(WhisperSTTService):
             import mlx_whisper
 
             await self.start_processing_metrics()
-            await self.start_ttfb_metrics()
 
             # Divide by 32768 because we have signed 16-bit data.
             audio_float = np.frombuffer(audio, dtype=np.int16).astype(np.float32) / 32768.0
@@ -413,7 +410,6 @@ class WhisperSTTServiceMLX(WhisperSTTService):
             if len(text.strip()) == 0:
                 text = None
 
-            await self.stop_ttfb_metrics()
             await self.stop_processing_metrics()
 
             if text:

From d2ac9006a25639ecf8326648b368319f027e2841 Mon Sep 17 00:00:00 2001
From: Mark Backman <mark@daily.co>
Date: Sat, 24 Jan 2026 12:49:52 -0500
Subject: [PATCH 51/51] Update env var to PIPECAT_SMART_TURN_LOG_DATA

---
 changelog/3525.added.md                                  | 2 +-
 src/pipecat/audio/turn/smart_turn/local_smart_turn_v3.py | 2 +-
 2 files changed, 2 insertions(+), 2 deletions(-)

diff --git a/changelog/3525.added.md b/changelog/3525.added.md
index 0dcf74287..e34b5e3d5 100644
--- a/changelog/3525.added.md
+++ b/changelog/3525.added.md
@@ -1 +1 @@
-- Added new `SMART_TURN_LOG_DATA` environment variable, which causes Smart Turn input data to be saved to disk
+- Added new `PIPECAT_SMART_TURN_LOG_DATA` environment variable, which causes Smart Turn input data to be saved to disk
diff --git a/src/pipecat/audio/turn/smart_turn/local_smart_turn_v3.py b/src/pipecat/audio/turn/smart_turn/local_smart_turn_v3.py
index 84f6b5622..1eae7cc02 100644
--- a/src/pipecat/audio/turn/smart_turn/local_smart_turn_v3.py
+++ b/src/pipecat/audio/turn/smart_turn/local_smart_turn_v3.py
@@ -49,7 +49,7 @@ class LocalSmartTurnAnalyzerV3(BaseSmartTurn):
         """
         super().__init__(**kwargs)
 
-        self._log_data = env_truthy("SMART_TURN_LOG_DATA", default=False)
+        self._log_data = env_truthy("PIPECAT_SMART_TURN_LOG_DATA", default=False)
 
         if not smart_turn_model_path:
             # Load bundled model