Agent skill

coderabbit-review

Use this skill when working with CodeRabbit, such as running CodeRabbit reviews, generating and processing automated CodeRabbit comments, or evaluating CodeRabbit suggestions.

Stars 163
Forks 31

Install this agent skill to your Project

npx add-skill https://github.com/majiayu000/claude-skill-registry/tree/main/skills/data/running-coderabbit-reviews

SKILL.md

CodeRabbit Review Handler

Process CodeRabbit review comments with context-aware discretion.

Instructions

1. Initial Context

Inform the user:

I'll review the CodeRabbit comments with discretion, as CodeRabbit doesn't have access to the entire codebase and may not understand the full context.

For each comment, I'll:
- Evaluate if it's valid given our codebase context
- Accept suggestions that improve code quality
- Ignore suggestions that don't apply to our architecture
- Explain my reasoning for accept/ignore decisions

2. Run Code Rabbit

If no comments are provided, run CodeRabbit in prompt-only mode to generate comments:

bash
# Run CodeRabbit cli
coderabbit --prompt-only 

3. Evaluate Comments

  • Parse the comments
  • For each comment:
    • Read the relevant file to understand context
    • Determine if the suggestion is valid and beneficial
    • Decide to accept or ignore the suggestion
    • Document reasoning for each decision
  • Ignore *.md files

4. Execute Fixes

Task a subagent to address the comments you've accepted

5. Validate Changes

Run relevant tests and linters to ensure code integrity:

Example:

bash
# Run all unit tests
./scripts/test.sh unit 1 2>&1 | tail -100

# Run affected UI tests
./scripts/test.sh ui [TestName] 2>&1 | tail -100

# Run swiftlint on changed files and fix ANY issues
git diff --name-only origin/main...HEAD -- '*.swift' | xargs -r swiftlint lint --strict

6. Commit Changes

After applying changes commit your work using appropriate commit messages summarizing the changes made. Address any re-commit hook violations as needed.

7. Consolidate Results

After completion, provide a summary report:

📋 CodeRabbit Review Summary

Files Processed: {count}

Accepted Suggestions:
  {file}: {changes_made}
  
Ignored Suggestions:
  {file}: {reason_ignored}

Overall: {X}/{Y} suggestions applied

Common Patterns to Ignore

  • Style preferences that conflict with project conventions
  • Generic best practices that don't apply to our specific use case
  • Performance optimizations for code that isn't performance-critical
  • Accessibility suggestions for internal tools
  • Security warnings for already-validated patterns
  • Import reorganization that would break our structure

Common Patterns to Accept

  • Actual bugs (null checks, error handling)
  • Security vulnerabilities (unless false positive)
  • Resource leaks (unclosed connections, memory leaks)
  • Type safety issues (TypeScript/type hints)
  • Logic errors (off-by-one, incorrect conditions)
  • Missing error handling

Decision Framework

For each suggestion, consider:

  1. Is it correct? - Does the issue actually exist?
  2. Is it relevant? - Does it apply to our use case?
  3. Is it beneficial? - Will fixing it improve the code?
  4. Is it safe? - Could the change introduce problems?

Only apply if all answers are "yes" or the benefit clearly outweighs risks.

Important Notes

  • CodeRabbit is helpful but lacks context
  • Trust your understanding of the codebase over generic suggestions
  • Explain decisions briefly to maintain audit trail
  • Batch related changes for efficiency
  • Always run the relevant ui and unit tests, and/or create new tests to verify correctness after applying changes

Didn't find tool you were looking for?

Be as detailed as possible for better results