code-quality（代码质量）

Code Quality Analysis

Comprehensive code quality analysis skill covering style, complexity, dead code detection, and type checking.

Core Capabilities

Style Compliance

Check adherence to language-specific style guides:

Python (PEP8/Black):

• Line length (88 chars for Black, 79 for PEP8)
• Indentation (4 spaces)
• Import ordering (standard, third-party, local)
• Naming conventions (snake_case for functions/variables, PascalCase for classes)
• Whitespace usage

Style check commands:

# Black formatting check
black --check --diff .

# Flake8 linting
flake8 --max-line-length=88 .

# isort import sorting
isort --check-only --diff .

Complexity Metrics

Measure and report code complexity:

Cyclomatic Complexity:

• Number of independent paths through code
• Target: < 10 per function
• Warning: 10-20
• Critical: > 20

Cognitive Complexity:

• Mental effort to understand code
• Accounts for nesting depth and control flow

Function Length:

• Target: < 50 lines
• Warning: 50-100 lines
• Critical: > 100 lines

Nesting Depth:

• Target: < 4 levels
• Warning: 4-6 levels
• Critical: > 6 levels

Complexity analysis:

# Using radon for Python
radon cc . -a -s  # Cyclomatic complexity
radon mi .        # Maintainability index
radon hal .       # Halstead metrics

Dead Code Detection

Identify unused code elements:

Unused Imports:

# Python - using autoflake
autoflake --check --remove-all-unused-imports .

# Using flake8 with F401
flake8 --select=F401 .

Unused Variables:

• Local variables assigned but never read
• Function parameters ignored
• Class attributes never accessed

Unused Functions/Classes:

• Defined but never called
• Private methods not used internally
• Dead code branches (always false conditions)

Unreachable Code:

• Code after return/raise/break/continue
• Branches with impossible conditions
• Deprecated code still in codebase

Type Hint Analysis

Validate type annotations:

Missing Type Hints:

# Using mypy
mypy --strict .

# Check specific strictness levels
mypy --disallow-untyped-defs .
mypy --disallow-incomplete-defs .

Type Errors:

• Incompatible types in assignments
• Wrong argument types
• Missing return types
• Generic type issues

Type Coverage:

• Percentage of code with type annotations
• Target: > 80% coverage

Quality Metrics Dashboard

When analyzing a codebase, report:

Analysis Workflow

Full Quality Analysis

To perform comprehensive quality check:

1. Style check: Run linters and formatters
2. Complexity analysis: Calculate metrics
3. Dead code scan: Find unused elements
4. Type check: Validate annotations
5. Report generation: Summarize findings

Quick Quality Check

For rapid assessment of changes:

1. Get changed files from git diff
2. Run focused linting on changed files
3. Check complexity of modified functions
4. Validate types in touched code

Language-Specific Guidance

Python

Recommended tooling:

• Black (formatting)
• isort (import sorting)
• flake8 (linting)
• mypy (type checking)
• radon (complexity)
• vulture (dead code)

Configuration (pyproject.toml):

[tool.black]
line-length = 88
target-version = ['py311']

[tool.isort]
profile = "black"
line_length = 88

[tool.mypy]
python_version = "3.11"
strict = true

JavaScript/TypeScript

Recommended tooling:

• ESLint (linting)
• Prettier (formatting)
• TypeScript compiler (type checking)

Configuration (.eslintrc.json):

{
  "extends": ["eslint:recommended"],
  "rules": {
    "complexity": ["warn", 10],
    "max-depth": ["warn", 4],
    "max-lines-per-function": ["warn", 50]
  }
}

Common Quality Issues

High Priority Fixes

1. Unused imports: Remove immediately
2. Type errors: Fix type mismatches
3. High complexity: Refactor into smaller functions
4. Long functions: Extract logical blocks

Medium Priority

1. Style violations: Apply formatter
2. Missing type hints: Add annotations
3. Moderate complexity: Consider refactoring
4. Unused variables: Remove or use

Low Priority

1. Minor style preferences: Team decision
2. Optional type hints: Add gradually
3. Documentation gaps: Address incrementally

Output Format

Present findings organized by severity:

Errors (must fix):

• Type errors
• Syntax issues
• Critical complexity

Warnings (should fix):

• Unused code
• High complexity
• Missing types in public APIs

Info (consider):

• Style suggestions
• Optimization opportunities
• Best practice recommendations

Integration

Coordinate with other skills:

• refactoring skill: For complexity reduction
• documentation skill: For missing docstrings
• security-scanning skill: For security-related quality issues