Update CHANGELOG

changelog/3175.added.md

@@ -10,14 +10,17 @@
      - `LLMThoughtStartFrame`
      - `LLMThoughtTextFrame`
      - `LLMThoughtEndFrame`
-  3. A generic mechanism for associating extra LLM-specific data with a function
+  3. A mechanism for appending arbitrary context messages after a function call
-     call in context, used specifically to support Google's function-call-related
+     message, used specifically to support Google's function-call-related
      "thought signatures", which are necessary to ensure thinking continuity
      between function calls in a chain (where the model thinks, makes a function
      call, thinks some more, etc.). See:
-     - `FunctionCallInProgressFrame.llm_specific_extra`
+     - `append_extra_context_messages` field in `FunctionInProgressFrame` and
-     - `LLMAssistantAggregator` handling of the above field
+       helper types
-     - `GeminiLLMAdapter` handling of`"tool_call_extra"` context messages
+     - `GoogleLLMService` leveraging the new mechanism to add a Google-specific
+       `"fn_thought_signature"` message
+     - `LLMAssistantAggregator` handling of `append_extra_context_messages`
+     - `GeminiLLMAdapter` handling of `"fn_thought_signature"` messages
   4. A generic mechanism for recording LLM thoughts to context, used
      specifically to support Anthropic, whose thought signatures are expected to
      appear alongside the text of the thoughts within assistant context
@@ -25,15 +28,14 @@
      - `LLMThoughtEndFrame.signature`
      - `LLMAssistantAggregator` handling of the above field
      - `AnthropicLLMAdapter` handling of `"thought"` context messages
-  5. A generic mechanism for recording standalone thought signatures to context,
+  5. Google-specific logic for inserting non-function-call-related thought
-     used specifically to support Gemini 3 Pro, which may return
+     signatures into the context, to help maintain thinking continuity in a
-     non-function-call-related thought signatures at the end of
+     chain of LLM calls. See:
-     non-function-call assistant responses. See:
+     - `GoogleLLMService` sending `LLMMessagesAppendFrame`s to add LLM-specific
-     - `LLMThoughtSignatureFrame`
+       `"non_fn_thought_signature"` messages to context
-     - `LLMAssistantAggregator` handling of the above frame
+     - `GeminiLLMAdapter` handling of `"non_fn_thought_signature"` messages
-     - `GeminiLLMAdapter` handling of `"thought_signature"` context messages
   6. An expansion of `TranscriptProcessor` to process LLM thoughts in addition
      to user and assistant utterances. See:
-     - `TranscriptProcessor.thought()`
+     - `TranscriptProcessor(process_thoughts=True)` (defaults to `False`)
-     - `ThoughtTranscriptionMessage`, which may now also be emitted with the
+     - `ThoughtTranscriptionMessage`, which is now also emitted with the
        `"on_transcript_update"` event