Problem Description

First, thank you for building agno — it's a fantastic framework.

I want to display the full agent execution trace that aprint_response provides (message, tool calls, reasoning, metrics, etc.) and also get the RunOutput object afterward — without making two LLM calls.

Previously this was possible via aprint_response + arun_response. Now that arun_response is removed, there's no clean way to do this.

What I've tried:

• arun(stream=True, yield_run_output=True) — gives the response object but loses the rich execution display
• InMemoryDb + aget_last_run_output() — works partially, but the content field is not deserialized back into the output_schema type (it comes back as a raw dict/string). Also, requiring a DB instance just to retrieve a result the agent already computed internally feels wrong.

Proposed Solution

aprint_response already holds run_response internally — it just doesn't return it. Returning it would solve this cleanly with no API breakage.

Alternatives Considered

No response

Additional Context

No response

Would you like to work on this?

• Yes, I’d love to work on it!
• I’m open to collaborating but need guidance.
• No, I’m just sharing the idea.