Add Dify integration and enhance workflow node specifications

- Introduce new fields `dify_api_url` and `dify_api_key` in `AssistantConfig` for Dify API integration. - Update `requirements.txt` to include `dify-client-python` for Dify SDK support. - Modify `config_resolver` to handle Dify connection information. - Add a new `globalNode` type in workflow specifications to provide unified settings across workflows. - Enhance node specifications with additional constraints and default values for better configuration management. - Update frontend components to support the new `globalNode` type and its properties, improving workflow editor functionality.
2026-07-11 22:26:31 +08:00
parent dfb9c5bd11
commit 00270a5c01
23 changed files with 1270 additions and 414 deletions
--- a/backend/services/brains/prompt_brain.py
+++ b/backend/services/brains/prompt_brain.py
@@ -0,0 +1,106 @@
+"""Local prompt assistant, including prompt-only reusable tools."""
+
+from __future__ import annotations
+
+from uuid import uuid4
+
+from models import AssistantConfig
+from pipecat.adapters.schemas.function_schema import FunctionSchema
+from pipecat.frames.frames import OutputTransportMessageUrgentFrame, TTSSpeakFrame
+from pipecat.processors.aggregators.llm_context import LLMContext
+from pipecat.processors.frame_processor import FrameProcessor
+from pipecat.services.llm_service import (
+    FunctionCallParams,
+    FunctionCallResultProperties,
+)
+from pipecat.utils.time import time_now_iso8601
+
+from services.brains.base import BaseBrain, BrainRuntime, BrainSpec
+
+
+class PromptBrain(BaseBrain):
+    spec = BrainSpec(
+        type="prompt",
+        supported_runtime_modes=frozenset({"pipeline", "realtime"}),
+        owns_context=True,
+    )
+
+    def build_llm(self, cfg: AssistantConfig, context: LLMContext) -> FrameProcessor:
+        from services.pipecat.service_factory import create_llm
+
+        return create_llm(cfg)
+
+    async def setup(self, cfg: AssistantConfig, runtime: BrainRuntime) -> None:
+        schemas: list[FunctionSchema] = []
+        for tool in cfg.tools:
+            if tool.type != "end_call":
+                continue
+            schema, handler = self._make_end_call_tool(tool, runtime)
+            schemas.append(schema)
+            runtime.llm.register_function(tool.function_name, handler)
+        runtime.set_tools(schemas)
+
+    @staticmethod
+    def _make_end_call_tool(tool, runtime: BrainRuntime):
+        config = (tool.definition or {}).get("config") or {}
+        message_type = str(config.get("message_type") or "none")
+        custom_message = str(config.get("custom_message") or "").strip()
+        capture_reason = bool(config.get("capture_reason", True))
+
+        async def end_call(params: FunctionCallParams) -> None:
+            reason = str(params.arguments.get("reason") or "end_call_tool").strip()
+            runtime.call_end.begin(reason)
+            await params.result_callback(
+                {"status": "success", "action": "ending_call"},
+                properties=FunctionCallResultProperties(run_llm=False),
+            )
+
+            if message_type != "custom" or not custom_message:
+                await runtime.call_end.finish()
+                return
+
+            turn_id = uuid4().hex
+            timestamp = time_now_iso8601()
+            for message in (
+                {
+                    "type": "assistant-text-start",
+                    "turn_id": turn_id,
+                    "timestamp": timestamp,
+                },
+                {
+                    "type": "assistant-text-delta",
+                    "turn_id": turn_id,
+                    "delta": custom_message,
+                },
+                {
+                    "type": "assistant-text-end",
+                    "turn_id": turn_id,
+                    "content": custom_message,
+                    "interrupted": False,
+                },
+            ):
+                await runtime.queue_frame(
+                    OutputTransportMessageUrgentFrame(message=message)
+                )
+            runtime.call_end.arm_after_speech()
+            await runtime.queue_frame(
+                TTSSpeakFrame(custom_message, append_to_context=False)
+            )
+
+        properties = (
+            {
+                "reason": {
+                    "type": "string",
+                    "description": "结束本次通话的简短原因。",
+                }
+            }
+            if capture_reason
+            else {}
+        )
+        schema = FunctionSchema(
+            name=tool.function_name,
+            description=tool.description or "结束当前通话。",
+            properties=properties,
+            required=["reason"] if capture_reason else [],
+        )
+        return schema, end_call