quality-check

Quality Check

Context

• Project type: !ls package.json 2>/dev/null && echo "node" || (ls setup.py pyproject.toml 2>/dev/null && echo "python" || echo "unknown")
• File count: !find src -type f \( -name "*.js" -o -name "*.ts" -o -name "*.py" \) 2>/dev/null | wc -l

Your Task

Analyze code quality and provide actionable recommendations for improvement.

Arguments

• $1: Analysis type (optional, default: "all")
  • smells - Detect code smells only
  • complexity - Analyze complexity only
  • all - Run all analyses
• $2: Severity threshold (optional, default: "medium")
  • low - Report all issues
  • medium - Report medium and high severity
  • high - Report only high severity issues

Steps

1. Detect project type and locate source files

2. Based on analysis type ($1):

   For code smells:

   • Scan for long methods (>50 lines)
   • Identify god classes (>20 methods)
   • Detect duplicated code
   • Find magic numbers
   • Check for deep nesting (>4 levels)

   For complexity:

   • Calculate cyclomatic complexity
   • Identify functions with complexity >10
   • Analyze cognitive complexity
   • Find deeply nested code

3. Categorize findings by severity

4. Filter by threshold ($2)

5. Provide refactoring recommendations

Output Format

# Code Quality Report

**Project:** [name]
**Files Analyzed:** [count]
**Issues Found:** [count]

## Summary

| Severity | Code Smells | Complexity | Total |
|----------|-------------|------------|-------|
| High     | X           | X          | X     |
| Medium   | X           | X          | X     |
| Low      | X           | X          | X     |

## High Priority Issues

### God Class: UserManager (src/services/UserManager.js)
- **Problem:** 523 lines, 18 methods, too many responsibilities
- **Impact:** Hard to maintain, test, and understand
- **Recommendation:** Split into UserService, AuthService, EmailService
- **Effort:** Medium (2-3 days)

### High Complexity: processOrder() (src/orders/processor.js:45)
- **Problem:** Cyclomatic complexity 28, cognitive complexity 35
- **Impact:** Error-prone, hard to test
- **Recommendation:** Extract validation, calculation, persistence
- **Effort:** Low (4-6 hours)

## Recommendations

1. **Immediate:** Refactor processOrder() - high complexity, high risk
2. **This Sprint:** Split UserManager god class
3. **Next Sprint:** Address duplicated code in payment processing