create-review-checklist

Generate review checklists for different change types. Use to customize review focus based on what was changed.

⭐6stars🔀1forks📁View on GitHub🕐Updated Dec 11, 2025

When & Why to Use This Skill

This Claude skill automates the generation of tailored code review checklists by analyzing the specific types of changes in a Pull Request. It helps developers and teams maintain high code quality by providing targeted verification steps for Python, Mojo, documentation, tests, and configuration files, ensuring no critical check is overlooked and improving overall review consistency.

Use Cases

  • Standardizing review processes across multi-language projects (e.g., Python and Mojo) to ensure consistent quality and style standards.
  • Providing structured guidance for developers reviewing unfamiliar change types or complex, multi-file pull requests.
  • Ensuring that non-code changes, such as documentation and configuration updates, receive appropriate scrutiny and validation alongside implementation code.
  • Streamlining large-scale code reviews by categorizing changes and prioritizing critical security, performance, and memory safety checks.
namecreate-review-checklist
description"Generate review checklists for different change types. Use to customize review focus based on what was changed."
categoryreview
mcp_fallbacknone
agenttest-engineer
user-invocablefalse

Create Review Checklists

Generate customized review checklists based on type of change.

When to Use

  • Starting PR review for unfamiliar change type
  • Need specific checklist for Mojo vs Python code
  • Reviewing test changes
  • Documenting-focused changes
  • Large multi-type changes needing organized review

Quick Reference

# Detect change type from PR
gh pr diff <pr> --name-only | grep -E "\\.mojo$|\\.py$|\\.md$|test_"

# Categorize changes
gh pr diff <pr> | head -100 | grep "^[+-]" | wc -l  # Changed lines

# Get file counts by type
gh pr diff <pr> --name-only | sed 's/.*\.//' | sort | uniq -c

Change Type Detection

Mojo Implementation:

  • Files: *.mojo or *.🔥
  • Focus: Syntax, patterns, memory safety, performance
  • Extra checks: SIMD, ownership, zero-warnings

Python Code:

  • Files: *.py
  • Focus: Style (PEP 8), type hints, error handling
  • Extra checks: Security, test coverage

Tests:

  • Files: test_*.py, test_*.mojo
  • Focus: Coverage, assertions, edge cases, clarity
  • Extra checks: Flakiness, isolation

Documentation:

  • Files: *.md, docstrings, comments
  • Focus: Clarity, accuracy, completeness
  • Extra checks: Links work, examples valid

Configuration:

  • Files: *.toml, *.yaml, .yaml, *.json
  • Focus: Correctness, consistency, validation
  • Extra checks: No secrets, proper syntax

Checklist Templates

Mojo Implementation Checklist:

  • v0.26.1+ syntax (no inout, @value, DynamicVector)
  • All __init__ use out self
  • Non-copyable returns use ^
  • Traits conformance correct (Copyable, Movable)
  • Memory safety validated
  • Zero compiler warnings
  • SIMD used in hot paths
  • Tests present and passing
  • Documentation updated

Python Code Checklist:

  • Follows PEP 8 style guide
  • Type hints on all functions
  • Docstrings present and clear
  • Error handling appropriate
  • No security vulnerabilities
  • Tests cover new code
  • Edge cases handled
  • No code duplication

Test Code Checklist:

  • Test name describes what's tested
  • Assertions are clear and specific
  • Edge cases covered (boundaries, empty, null, large)
  • Setup/teardown clean (no side effects)
  • Not dependent on other tests
  • Reasonable timeout (not too long)
  • Mocking/isolation appropriate
  • Deterministic (no randomness/flakiness)

Documentation Checklist:

  • Spelling and grammar correct
  • Links validated and working
  • Code examples are complete and correct
  • Instructions tested and accurate
  • Structure is logical and easy to follow
  • Markdown formatting valid
  • No broken references
  • Up to date with code changes

Configuration Checklist:

  • Syntax is valid (YAML, TOML, JSON)
  • No hardcoded secrets
  • Consistent with project standards
  • Required fields present
  • Default values sensible
  • Documentation matches config
  • Backward compatible (if applicable)
  • Performance impact acceptable

Checklist Generation Workflow

  1. Analyze PR: Determine file types changed
  2. Categorize: Group changes by type
  3. Select templates: Pick appropriate checklists
  4. Customize: Adjust based on complexity/scope
  5. Prioritize: Mark critical vs optional items
  6. Document: Create checklist with explanations
  7. Use: Apply during code review

Output Format

Report checklist with:

  1. Change Summary - What types of changes detected
  2. Primary Focus - Main review area
  3. Checklist Items - Organized by category
  4. Deep Dives - Detailed items for complex changes
  5. Quick Wins - Easy items to verify first
  6. Risks - High-risk areas to focus on
  7. Notes - Special considerations

Multi-Type Example

PR changing Mojo code + Tests + Docs:

  1. Start with Mojo checklist (primary)
  2. Apply test checklist to test changes
  3. Apply documentation checklist to docs
  4. Verify consistency across all types
  5. Check integration between parts

Customization Rules

  • Critical items must PASS before approval
  • High items SHOULD pass unless justified
  • Medium items are NICE to have
  • Low items are OPTIONAL suggestions

Critical (must fix):

  • Syntax errors
  • Test failures
  • Security issues
  • Breaking changes

High (should fix):

  • Code style issues
  • Missing tests
  • Performance regression
  • Incomplete documentation

Medium (nice to have):

  • Code cleanup
  • Example improvements
  • Comment refinements

Low (optional):

  • Formatting polish
  • Minor optimizations
  • Documentation color

Error Handling

Problem Solution
Mixed change types Create separate checklists for each type
Unclear type Inspect files to determine type
Complex change Break into multiple checklists
Specialized domain Add domain-specific items to template
New pattern Create new checklist template

References

  • See review-pr-changes for full review workflow
  • See CLAUDE.md for code standards
  • See individual skill docs for detailed requirements