wezell/Codex Code Review & Actor
Run Codex to review a PR and post review comments
View on GitHubTrust Signals
- Scorecard Score
- not yet scored
- Maintenance Recency
- Stale
- License
- None
Inputs
| name | description | required | default |
|---|---|---|---|
| mode | Operation mode: review | act | no | review |
| openai_api_key | OpenAI API key for the selected model provider | no | — |
| openrouter_api_key | OpenRouter API key used for all routed model calls | no | — |
| config_path | Path (relative to the consuming repo's checkout) to the in-repo OpenRouter model config file. Edit that file to swap review or act model — no code change required. Defaults to .openrouter-review.yml at the repo root. | no | .openrouter-review.yml |
| model | Model to use (e.g., gpt-5, gpt-4.1-mini) | no | gpt-5.4 |
| reasoning_effort | Reasoning effort: minimal | low | medium | high | no | medium |
| web_search_mode | Web search mode: disabled | cached | live | no | live |
| debug_level | Debug verbosity: 0 (off), 1 (basic), 2 (trace) | no | 1 |
| dry_run | If '1', print payloads but do not post comments | no | 0 |
| stream_agent_messages | If '1', stream model output to logs | no | 1 |
| act_instructions | Additional instructions for act mode (autonomous editing) | no | "" |
| allowed_commenter_associations | Comma-separated GitHub comment author_association values allowed to trigger act mode | no | MEMBER,OWNER,COLLABORATOR |
| additional_prompt | Extra reviewer instructions appended to the review prompt (verbatim) | no | "" |
| extra_pip_args | Additional pip arguments (e.g., --index-url) | no | "" |
Outputs
no outputs