modules/programs/agents/shared/skills/codex-review/SKILL.md
Send staged changes to Codex for independent peer review, then evaluate and implement warranted feedback
npx skillsauth add MichaelVessia/nixos-config codex-reviewInstall this skill globally with one command. Works with Claude Code, Cursor, and Windsurf.
3 of 9 scanners reported clean
Some scanners were skipped, did not run, or reported a non-clean status. Review each row below.
Use the codex subagent to review staged changes. Pass it the following prompt:
Review the staged git diff as a senior engineer. Provide:
- Critical Issues - bugs, security vulnerabilities, or breaking changes that MUST be fixed.
- Improvements - concrete suggestions for better code quality, performance, or maintainability.
- Nitpicks - minor style or preference items (low priority).
Be specific. Reference exact file paths and lines from the diff. Provide code examples for suggested fixes. Do NOT comment on formatting or whitespace. Do NOT suggest adding comments or docblocks unless something is genuinely confusing. Keep your review concise and actionable.
Display Codex's full review to the user in a clear, formatted way.
Read the staged diff yourself (git diff --cached) so you have full context,
then evaluate each piece of Codex's feedback:
After making changes, re-stage the modified files.
development
Generate self-contained HTML visualizations with Plannotator theming. Use for implementation plans, PR explainers, architecture diagrams, data tables, slide decks, and any visual explanation of technical concepts. Plans and PR explainers follow Plannotator's prescriptive approach; all other visual content delegates to nicobailon/visual-explainer.
development
Turn an idea or objective into a goal package for /goal. Interviews the user, builds a reviewed fact sheet via Plannotator, then explores the codebase to produce an execution plan.
development
Open Plannotator's browser-based code review UI for the current worktree or a pull request URL, then act on the feedback that comes back.
testing
Open Plannotator on the latest rendered assistant message and use the returned annotations to revise that message or continue.