🤖 feat: mark output reserve on auto-compact sliders #1760
Conversation
![chatgpt-codex-connector[bot]](https://avatars.githubusercontent.com/in/1144995?s=60&v=4){kind=small}
There was a problem hiding this comment.
💡 Codex Review
Here are some automated review suggestions for this pull request.
Reviewed commit: abf6a58230
ℹ️ About Codex in GitHub
Codex has been enabled to automatically review pull requests in this repo. Reviews are triggered when you
- Open a pull request for review
- Mark a draft as ready
- Comment "@codex review".
If Codex has suggestions, it will comment; otherwise it will react with 👍.
When you sign up for Codex through ChatGPT, Codex can also answer questions or update the PR, like "@codex address that feedback".
|
@codex review Resolved review feedback (removed useMemo; avoided prop destructuring). |
|
Codex Review: Didn't find any major issues. Nice work! ℹ️ About Codex in GitHubCodex has been enabled to automatically review pull requests in this repo. Reviews are triggered when you
If Codex has suggestions, it will comment; otherwise it will react with 👍. When you sign up for Codex through ChatGPT, Codex can also answer questions or update the PR, like "@codex address that feedback". |
ammar-agent
force-pushed
the
context-usage-c6p2
branch
2 times, most recently
from
a48e7bf to
9b93b67
Compare
- Remove hardcoded 32000 default in streamManager.ts - Use model's max_output_tokens from metadata when caller doesn't specify - If no metadata exists, omit parameter entirely (let provider default) - Clarify suggestion.ts comment that we use max_input_tokens for comparisons
ammar-agent
force-pushed
the
context-usage-c6p2
branch
from
48d55a2 to
c23e45a
Compare
2 participants
Summary
Why
The UI needs to reflect the real output reserve boundary so users don’t set auto-compact above a hidden output cap and hit
context_exceededfirst. DRYing the logic keeps the sidebar + status button in sync, and the warning text needed higher contrast in light theme to be readable.Testing
Generated with
mux• Model:openai:gpt-5.2-codex• Thinking:high• Cost:$26.67