AgentSkillsCN

detect-code-smells

识别代码中的质量问题与反模式，帮助在代码评审中发现潜在的可维护性隐患。

SKILL.md
--- frontmatter
name: detect-code-smells
description: "Identify code quality issues and anti-patterns. Use when reviewing code for maintainability problems."
mcp_fallback: none
category: analysis
tier: 2

Detect Code Smells

Identify code quality issues, anti-patterns, and maintainability problems that suggest need for refactoring.

When to Use

  • •Code review before merging
  • •Identifying refactoring priorities
  • •Mentoring and code quality improvement
  • •Planning technical debt reduction

Quick Reference

bash
# Python code smell detection
pip install pylint radon
pylint --disable=all --enable=convention,refactor module.py
radon cc -a module.py  # Cyclomatic complexity
radon mi -s module.py  # Maintainability index

Workflow

  1. •Scan for patterns: Identify common code smell patterns (long functions, duplication, magic numbers)
  2. •Check complexity: Measure cyclomatic and cognitive complexity
  3. •Evaluate naming: Check for unclear variable/function names
  4. •Review structure: Identify violations of SOLID principles
  5. •Document findings: List smells with severity and recommendations

Output Format

Code quality report:

  • •Code smell type (duplication, long method, magic number, etc.)
  • •Location (file:line)
  • •Severity (critical/high/medium/low)
  • •Explanation of the issue
  • •Suggested fix or refactoring approach

References

  • •See CLAUDE.md > SOLID principles for design guidance
  • •See CLAUDE.md > DRY and KISS principles for quality standards
  • •See refactor-code skill for applying fixes