Added a new LLMAssistantPushAggregationFrame control frame that signals LLMAssistantAggregator to immediately flush its text buffer to the conversation context

src/pipecat/frames/frames.py

@@ -1990,6 +1990,16 @@ class LLMFullResponseEndFrame(ControlFrame):
         self.skip_tts = None
 
 
+@dataclass
+class LLMAssistantPushAggregationFrame(ControlFrame):
+    """Frame that forces the LLM assistant aggregator to push its current aggregation to context.
+
+    When received by ``LLMAssistantAggregator``, any text that has been accumulated
+    in the aggregation buffer is immediately committed to the conversation context as
+    an assistant message, without waiting for an ``LLMFullResponseEndFrame``.
+    """
+
+
 @dataclass
 class LLMContextSummaryRequestFrame(ControlFrame):
     """Frame requesting context summarization from an LLM service.

src/pipecat/processors/aggregators/llm_response_universal.py

@@ -35,6 +35,7 @@ from pipecat.frames.frames import (
     InputAudioRawFrame,
     InterimTranscriptionFrame,
     InterruptionFrame,
+    LLMAssistantPushAggregationFrame,
     LLMContextAssistantTimestampFrame,
     LLMContextFrame,
     LLMContextSummaryRequestFrame,
@@ -879,6 +880,8 @@ class LLMAssistantAggregator(LLMContextAggregator):
         elif isinstance(frame, (EndFrame, CancelFrame)):
             await self._handle_end_or_cancel(frame)
             await self.push_frame(frame, direction)
+        elif isinstance(frame, LLMAssistantPushAggregationFrame):
+            await self.push_aggregation()
         elif isinstance(frame, LLMFullResponseStartFrame):
             await self._handle_llm_start(frame)
         elif isinstance(frame, LLMFullResponseEndFrame):