| name | parallel-review |
| description | Run code review and coderabbit review in parallel on changed code. Use when user says "review", "code review", "run review", "check my code", "review changes", or after completing a task/feature. Also trigger proactively after significant code changes. |
Parallel Code Review
Runs multiple code review tools simultaneously for comprehensive feedback.
Workflow
Step 1: Determine Scope
Figure out what to review:
- If user specifies files → review those files
- Otherwise → review all unstaged/staged changes via
git diff
Step 2: Launch Parallel Reviews
Run ALL of the following in parallel using the Agent tool:
-
Code Review Agent (subagent_type: "pr-review-toolkit:code-reviewer")
- Prompt: "Review the following changed files for bugs, logic errors, code quality, and adherence to project CLAUDE.md conventions: [list files or describe changes]"
-
CodeRabbit Review Agent (subagent_type: "coderabbit:code-reviewer")
- Prompt: "Review the code changes for bugs, security issues, and quality problems: [list files or describe changes]"
-
Silent Failure Hunter (subagent_type: "pr-review-toolkit:silent-failure-hunter")
- Only include this if the changes involve error handling, try-catch, or fallback logic
- Prompt: "Check for silent failures and inadequate error handling in: [list files]"
Step 3: Synthesize Results
After all agents complete:
- Combine findings, deduplicating overlapping issues
- Prioritize: Critical > Serious > Moderate > Minor
- Present a unified summary with actionable fixes
- If issues are found, offer to fix them
Step 4: Fix (Optional)
If user agrees to fixes:
- Apply fixes for critical and serious issues
- Re-run
/fix-ts-errors on changed files
- Briefly confirm what was fixed
For UI Changes
If the changes include frontend/UI code, also run in parallel:
/web-interface-guidelines review/ui-skills review/rams accessibility review
Quick Mode
If user says "quick review" — run only the code-reviewer agent (skip coderabbit and silent-failure-hunter).
