Use when adding new error messages to React, or seeing "unknown error code" warnings.
1
0
# Install this skill:
npx skills add jcdoll/agents --skill "code-review"
Install specific skill from multi-skill repository
# Description
>-
# SKILL.md
name: code-review
description: >-
Reviews code for quality, security, and maintainability. Use after
implementing features, completing significant work, or when user asks for code
review.
allowed-tools:
- Read
- Grep
- Glob
- Bash
Code Review
Instructions
- Run
git diffto identify changed files - Read modified files to understand the intent of changes
- Review against the checklist below
- Provide structured feedback organized by priority
Review Checklist
Security
- No exposed credentials, API keys, or secrets
- Input validation implemented where needed
- File paths are sanitized (no path traversal)
- External commands are safely constructed (no injection)
Code Quality
- Variables and functions are clearly named
- Functions have single responsibility
- No code duplication (DRY principle)
- Complex logic has explanatory comments
- Error handling is comprehensive
- No unused imports or variables
- Clean separation of concerns
Parameters
- Use enums rather than bare strings for parameters
- Do not spread "magic strings" throughout the code, use a single source of truth for parameter defaults (config.py)
- Do not use the "if (check config) then (value = from config) else (value = default)" pattern - this breaks the single source of truth rule
Python Specific
- Type hints used appropriately
- Exceptions are specific (not bare
except:) - Context managers used for resources
- f-strings preferred over
.format()or%
Testing
- New functionality has test coverage
- Edge cases are tested
- Error conditions are tested
- Tests are readable and maintainable
Project Standards
- Code passes
rufflinting - No legacy wrappers or thin compatibility layers
- No "last updated" dates or copyright headers
Output Format
Organize feedback by severity:
Critical (must fix)
- Security issues
- Bugs that will cause failures
- Include specific line numbers and suggested fixes
Warnings (should fix)
- Code quality issues
- Missing error handling
- Potential edge cases
- Include rationale and improvement suggestions
Suggestions (consider)
- Style improvements
- Minor optimizations
- Include brief explanation of benefit
# Supported AI Coding Agents
This skill is compatible with the SKILL.md standard and works with all major AI coding agents:
Amp
Antigravity
Claude Code
Clawdbot
Codex
Cursor
Droid
Gemini CLI
GitHub Copilot
Goose
Kilo Code
Kiro CLI
OpenCode
Roo Code
Trae
Windsurf
Learn more about the SKILL.md standard and how to use these skills with your preferred AI coding agent.