| name | review |
| description | Thorough code review skill for checking your work before completion. Use this after implementing a feature or fix to verify code quality, completeness, testing, and documentation. |
Code Review
Perform a thorough code review of your changes before marking your task as complete.
When to Use
Run /review after you have finished implementing a feature, bug fix, or any code changes, but BEFORE running fab issue close or fab agent done.
How to Review
Use the Task tool to spawn a sub-agent for code review. This provides a fresh perspective without the context of how the code was written, leading to more objective reviews.
Step 1: Run the Sub-Agent Review
Use the Task tool with subagent_type: "general-purpose" and a prompt like:
Review all code changes (committed and uncommitted) between main and the current worktree. Run `git diff main` to see all changes.
Check each of these areas:
1. **Correctness**: Does the code solve the problem? Are there logic errors or unhandled edge cases?
2. **Completeness**: Is the implementation fully complete? Any TODOs left behind?
3. **Testing**: Are there tests for new functionality? Do existing tests pass?
4. **Code Quality**: Is the code readable and well-organized? Does it follow project conventions?
5. **Documentation**: Are public APIs documented? Do complex algorithms have comments?
6. **Security**: Any potential vulnerabilities (injection, XSS, etc.)? Is user input validated?
7. **Performance**: Any obvious performance issues (N+1 queries, unnecessary loops)?
Run the test suite and any linters configured for the project.
Report:
- Issues found (with file paths and line numbers)
- Suggestions for improvement
- Confidence level that the implementation is complete and correct
Step 2: Address Feedback
CRITICAL: You MUST address all issues found during review before proceeding. Do not skip, defer, or ignore feedback. The review exists to catch problems - ignoring it defeats the purpose.
After the sub-agent completes its review:
- Carefully read through ALL issues and suggestions reported
- Fix every problem identified - no exceptions
- If the sub-agent raised concerns, address each one explicitly
- If significant changes were made, run
/review again to verify fixes
Step 3: Verify Fixes
Run the test suite one more time to ensure your fixes didn't introduce new issues.
Output
After completing the review process, summarize:
- What the sub-agent found
- What you fixed (be specific - list each fix)
- Confidence level that the implementation is complete and ready
IMPORTANT: You may NOT proceed to commit or close the issue until ALL review feedback has been addressed. If the review found issues, you must fix them first. Do not proceed with unresolved feedback.
