[Feature]: Add a more lenient timeout mechanism to the gpt-*-xhigh models for custom channels #1925
Description
Prerequisites
- I will write this issue in English (see our Language Policy)
- I have searched existing issues and discussions to avoid duplicates
- This feature request is specific to oh-my-opencode (not OpenCode core)
- I have read the documentation or asked an AI coding agent with this project's GitHub URL loaded and couldn't find the answer
Problem Description
When omo invokes a sub-agent and uses the GPT model with the xhigh variant, it often requires a long time to think and read files without making any changes. After more than 10 minutes, omo considers this a dead agent and chooses to delegate to another sub-agent or take over the task itself.
Proposed Solution
For models with xhigh variants, if actual output or tool invocation records are detected, a more relaxed timeout mechanism should be added.
Alternatives Considered
I tried entering in advance in Atlas: the backend sub-agent is taking a long time to think without making changes or the changes are not yet complete, please do not mark the task as completed or reassign it to a new agent or handle it yourself, but it didn't work.
Doctor Output (Optional)
Additional Context
The part marked with a red line means that the xhigh model has already exceeded 10 minutes of time—even though it is still working.
Feature Type
New Hook
Contribution
- I'm willing to submit a PR for this feature
- I can help with testing
- I can help with documentation