Skip to content
Discussion options

You must be logged in to vote

Hi @thebigcosinus, thanks for the question, and sorry it sat this long without a reply.

The logs you found in strands.multiagent.a2a.executor are the assistant's server side. What you actually want is the message the orchestrator sends, and that's easiest to capture on the orchestrator itself. Why? Because when the master calls an assistant through the A2A client tools, the outgoing message is just the tool input. A hook shows it directly:

from strands.hooks import BeforeToolCallEvent

def log_outgoing(event: BeforeToolCallEvent) -> None:
    print(f"tool: {event.tool_use['name']}, input: {event.tool_use['input']}")

orchestrator.add_hook(log_outgoing)

The matching AfterToolCallEvent give…

Replies: 1 comment

Comment options

You must be logged in to vote
0 replies
Answer selected by yonib05
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Category
Q&A
Labels
None yet
2 participants