-
Notifications
You must be signed in to change notification settings - Fork 6k
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
AssistantAgent
should have a more general way to trigger a StopMessage
response.
#4020
AssistantAgent
should have a more general way to trigger a StopMessage
response.
#4020
Comments
To be clear, the assistant agent uses the string to respond with a |
Yeah .. some notes here .. The AssistantAgent default prompt encourages it to respond with Terminate when the task is done hence the check for terminate and issuing a stop message. The system message and terminate logic are related here. For example assuming the user supplies their own system message, the termination logic has less of an effect. |
Yes, we can think of a more general way to trigger a |
AssistantAgent
should have a more general way to trigger a StopMessage
response.
Or maybe don't send it at all. Let the Team decide. |
|
Seems to be resolved in #4042 . |
Patterned off of the v0.2 AssistantAgent, some agents include very specific termination logic which might go against the Team or Orchestrator logic. E.g.,
autogen/python/packages/autogen-magentic-one/src/autogen_magentic_one/agents/coder.py
Line 35 in 4023454
We should standardize where termination logic lives.
The text was updated successfully, but these errors were encountered: