When using “team”, regardless of the mode, is it necessary for the leader to submit every member’s results to the large model for summarization before returning the final result? Is there a parameter or method to skip this step?
This is important because sometimes the data returned by the tools is very large, which can easily cause the model’s context length to be exceeded.
In agent calls to tools, there is a parameter called stop_after_tool_call_tools that seems to be designed exactly for this purpose. When set to True, the agent does not summarize the tool’s output.
Now, since agents are used inside the team, is there a similar configuration available in the team setup to prevent summarization of the agent outputs?
Hey @tom, thanks for reaching out and supporting Agno. I’ve shared this with the team, we’re working through all requests one by one and will get back to you soon.If it’s urgent, please let us know. We appreciate your patience!