Distributed multi agent deployment using handoffs

[JensCardon]

Hi authors,

First of all, love the work you've done, big props for making this open source!

I want to deploy a multi agent system using your repo where each agent is hosted behind a different end point. Is there a way to use the handoff feature using this setup?

Much appreciated

[rm-openai]

Yeah you can use the Handoff object. Something like:

async def _invoke_handoff(ctx: RunContextWrapper[Any], input_json: str | None = None) -> Agent[Any]:
   handoff_agent: Agent = await fetch_from_remote_server(...)
   return handoff_agent

main_agent = Agent(..., handoffs=[Handoff(tool_name=..., tool_description=...., on_invoke_handoff=_invoke_handoff)

You could also use agents as tools.

What's the use case for hosting agents on different servers?

[github-actions]

This issue is stale because it has been open for 7 days with no activity.

[github-actions]

This issue was closed because it has been inactive for 3 days since being marked as stale.