always-further avatar

code-review

Activates when user wants code reviewed for quality, best practices, bugs, or improvements. Triggers

提供方 always-further|开源

Code Review Skill

Overview

This skill enables thorough, professional code reviews that identify bugs, security issues, and opportunities for improvement while remaining constructive and educational.

Activation Triggers

This skill activates when you:

  • Ask for code to be reviewed
  • Want feedback on implementation
  • Need a quality check
  • Ask about best practices for specific code

Example prompts:

  • "Review this function for me"
  • "Is this implementation correct?"
  • "Find potential bugs in this code"
  • "How can I improve this class?"
  • "Check this for security issues"

Review Categories

Correctness

  • Logic errors
  • Edge cases
  • Null handling
  • Race conditions

Security

  • Input validation
  • Injection vulnerabilities
  • Authentication issues
  • Data exposure

Performance

  • Algorithm efficiency
  • Memory usage
  • Database queries
  • Caching

Maintainability

  • Code clarity
  • Naming conventions
  • DRY principle
  • SOLID principles

Example Usage

Example 1: Function Review

User: "Review this authentication function"

Claude: Analyzes the function for security vulnerabilities (password handling, timing attacks), correctness (edge cases, error handling), and best practices. Provides specific line-by-line feedback.

Example 2: Pull Request Review

User: "Review the changes in this PR"

Claude: Examines the diff, identifies issues, suggests improvements, and provides an overall assessment with clear approve/request changes recommendation.

Output Format

Reviews are structured as:

  1. Summary: Overall assessment
  2. Critical Issues: Must-fix problems
  3. Suggestions: Nice-to-have improvements
  4. Positive Aspects: What's done well

Author

Claude Code Community

Version History

  • 1.0.0: Initial release
code-review - 适用于 Claude Code 与 Cursor 的 AI 智能体 Skill | Agent Skills