MultiAgentWorkflow

[logan-markewich]

A multi agent system implemented as a prebuilt workflow.

Currently, supports

• state injection into latest user messages
• accessing workflow context/state through the FunctionToolWithContext
• human in the loop using the new ctx.wait_for_event()
• both react and function calling llms
• maintaining context between runs
• direct (and controllable) handoffs between agents

Usage Example:
https://gist.github.com/logan-markewich/f91929a4d0c7e41515bed1d2851c566c

Current Diagram:

Todo:

• dogfooding
• more tests
• improve docs?

[marcusschiesser]

are these agents streaming dedicated events that can be shown in the UI (we're using this in create-llama).

There we're having the AgentRunEvent see
https://github.com/run-llama/create-llama/blob/main/templates/components/multiagent/python/app/workflows/events.py
(ignore to_response - this is for conversion to vercel data streams, this concern can be done outside of this PR)

here is an example using it to send the progress of tool calls:
https://github.com/run-llama/create-llama/blob/eec237c5feea1af9cdd5b276d34ebe3b8d0fd185/templates/components/multiagent/python/app/workflows/tools.py#L141

[logan-markewich]

Yes! That is the main intention for these events, to show progress in some UI.

I didn't capture the concept of "in progress" or "completed" with this, its mostly all just events at points in time (here's the agent input, here's the agent stream, here's the agent output, heres a tool im about to call, here's the tool output) -- I could refactor, but not sure if its needed or not

[review-notebook-app]

Check out this pull request on 

See visual diffs & provide feedback on Jupyter Notebooks.

---

Powered by ReviewNB