v0.63.1

deepfakes/faceswapv0.63.1Jul 22, 2026by github-actions[bot]

AI Summary

This patch release focuses on improving the reliability of Composio task execution and tool-call parsing, specifically addressing recovery from malformed model outputs. It includes improvements to workflow rendering and chat control.

Key Highlights

Composio task execution now avoids a 'first call bounce' when arguments already match the contract.
Tool-call parsing has been hardened to recover from garbled Kimi-K2-family tool-call bodies.
Workflows copilot and canvas received a major reliability pass for consistent rendering.

Full Release Notes

# The Composio Reliability Upgrade

This patch release tightens up Composio task execution and makes tool-calls more resilient, so your agents can get to work faster and recover more gracefully when model output gets messy. 🚀

## Highlights

### Composio task execution & schema validation 🧩
Composio actions triggered via `integrations_agent` now avoid a frustrating “first call bounce” and can execute immediately when the model’s arguments already match the resolved live contract. This prevents loops like “fetch my latest email” stalling before `GMAIL_FETCH_EMAILS` ever runs, especially along the text-mode per-action path. ([#5132](https://github.com/tinyhumansai/openhuman/pull/5132)) — Thank you @M3gA-Mind!

### Tool-call parsing robustness (Kimi recovery) 🛠️
Tool-call parsing has been hardened to better recover from Kimi-K2-family garbled tool-call bodies. The normalization now strips the `<|\"|>` argument-quote sentinels and rewrites malformed `NAME{unquoted-json}` into a canonical form, improving resilience when model output is slightly corrupted. ([#5132](https://github.com/tinyhumansai/openhuman/pull/5132)) — Thank you @M3gA-Mind!

## Contributor Credits

Thanks to everyone who contributed to v0.63.1: @M3gA-Mind and @github-actions[bot].

## Full Compare

https://github.com/tinyhumansai/openhuman/compare/v0.63.0...v0.63.1