Conversation
Codex Review SummaryThis comment shows the latest Codex review activity on this pull request.
ℹ️ About Codex in GitHubYour team has set up Codex to review pull requests in this repo. Reviews are triggered when you
Codex reacts with 👀 while any review is running, comments if it has suggestions, and reacts with 👍 once all reviews finish with no findings. |
![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: f33a918240
ℹ️ About Codex in GitHub
Your team has set up Codex to 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 👍.
Codex can also answer questions or update the PR. Try commenting "@codex address that feedback".
| Companion defaults: model=CC_PLUGIN_CODEX_DEFAULT_MODEL (otherwise opus), and no effort. The companion forwards `--effort` only when the user passes it, so `fable`, `opus`, `sonnet`, and `haiku` each keep Claude Code's own effort default. Claude Code owns which effort levels each model supports. | ||
|
|
||
| When the user omits `--model`, omit it from the companion command so the configured default can apply. |
There was a problem hiding this comment.
When CC_PLUGIN_CODEX_DEFAULT_MODEL is configured and rescue delegates to its forwarding worker, internal-skills/cli-runtime/runtime.md:27 still states that the companion defaults to opus. This leaves the internal execution contract inconsistent with both this public skill and the shared resolver, so update that runtime reference to describe the environment-variable fallback as part of this contract change.
AGENTS.md reference: AGENTS.md:L5-L5
Useful? React with 👍 / 👎.
1 participant
Review, adversarial review, and rescue currently select
opuswhenever--modelis omitted, requiring users who prefer another model to repeat the flag on each invocation.Add
CC_PLUGIN_CODEX_DEFAULT_MODELat the shared default-model resolver, with precedence--model> environment setting >opus. Trim configured values and preserve the existing fallback for unset or blank values. The turn-end review gate continues to use Claude Code's own defaults.Document configuration through Codex's
shell_environment_policy.set. Update the public skills and internal runtime guide to omit--modelwhen the user omits it, so the configured default can apply. Tests cover default selection, explicit precedence, whitespace handling, and forwarding the resolved model into CLI arguments.Validation: the implementation previously passed
npm run check(517 unit, 45 integration, and 23 E2E tests), including lint, typecheck, version, and changelog checks. The runtime-guide follow-up passes all 12 skill-contract tests andgit diff --check. Tests ran with inherited desktop/plugin configuration isolated from the fixtures.