fix(langchain): set agent name on AIMessage responses when name parameter is provided #33778
+265
−3
Conversation
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
…vided - Added _set_agent_name_on_message helper function to set name on AIMessage objects - Updated _handle_model_output to accept and use agent_name parameter - Updated _execute_model_sync and _execute_model_async to pass name parameter - Added comprehensive test coverage (8 test cases) for agent name functionality Previously, the name parameter in create_agent() was only used for naming the compiled graph, not for setting AIMessage.name in responses. This fix ensures that when a name is provided, it is correctly reflected in all AIMessage responses from the agent. Fixes: langchain-ai#33754
…gent_name.py - Reformatted function parameters in _handle_model_output for better clarity. - Cleaned up whitespace in test cases to enhance readability. - Ensured consistent assertion formatting in tests for agent name verification. These changes aim to improve code maintainability and readability without altering functionality.
Comment on lines
+810
to
+818
| return AIMessage( | ||
| content=msg.content, | ||
| tool_calls=msg.tool_calls, | ||
| invalid_tool_calls=msg.invalid_tool_calls, | ||
| additional_kwargs=msg.additional_kwargs, | ||
| response_metadata=msg.response_metadata, | ||
| usage_metadata=msg.usage_metadata, | ||
| name=agent_name, | ||
| id=msg.id, |
There was a problem hiding this comment.
I don't think you need to create a new AIMessage object here. You could just add your validation on _handle_model_output and then assign the name directly with output.name = agent_name
There was a problem hiding this comment.
Yes, while this change reduces the number of lines, the main reason for it was to enforce a fixed structure, which improves overall maintainability, readability, and consistency across the codebase. Please let me know if I’ve misunderstood anything.
There was a problem hiding this comment.
That's fair, but i see other places where we are updating just one field from AIMessage and it's using direct assignment. However, it's just a nit, so up to you :)
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
2 participants
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
Description:
This PR fixes a bug where the
nameparameter provided tocreate_agent()was not being set onAIMessageobjects in the agent responses. Previously,AIMessage.namewas alwaysnullregardless of thenameparameter value, even though the parameter was documented to work.The fix:
_set_agent_name_on_message()helper function to set name on AIMessage objects_handle_model_output()to accept and useagent_nameparameter_execute_model_sync) and async (_execute_model_async) model execution paths to pass the name parameterThis ensures backward compatibility - when
nameis not provided,AIMessage.nameremainsNone.Issue:
Fixes the bug reported where
nameparameter increate_agent()was not reflected in AIMessage responses. (Add issue number if you have it)Dependencies:
None. No new dependencies required.