/refactor-scan [target]

Analyze code for refactoring opportunities. Read-only — no changes made.

No arguments? Describe this skill and stop. Do not execute.

On Invoke

1. Read target files (argument or recently discussed code)
2. Analyze against refactoring criteria
3. Report issues found with locations and severity
4. Show plan of what /refactor-check-fix would do

Criteria to Check

AI Antipatterns to Flag

• Over-abstraction (factories/wrappers used once)
• Defensive paranoia (impossible null checks)
• Comment spam (obvious comments)
• Speculative features (unused config options)
• Wrapper classes adding no value

Output Format

═══════════════════════════════════════════
 REFACTOR SCAN: [target]
═══════════════════════════════════════════

 Summary: X issues found (Y high, Z medium)

 HIGH SEVERITY
   [file:line] Function too long (45 lines)
     → Split into: parseInput(), validateData(), formatOutput()

   [file:line] File too long (387 lines)
     → Extract: utils.ts, validators.ts

 MEDIUM SEVERITY
   [file:line] Vague name: "data"
     → Rename to: userPreferences

   [file:line] Deep nesting (4 levels)
     → Flatten with early returns

 LOW SEVERITY
   [file:line] Magic number: 86400
     → Extract: SECONDS_PER_DAY = 86400

 AI ANTIPATTERNS
   [file:line] Over-abstraction: ConfigFactory used once
     → Inline the factory

═══════════════════════════════════════════
 REFACTOR PLAN
═══════════════════════════════════════════

 1. Split [file] into focused modules
 2. Extract [function] from [large-function]
 3. Rename [vague-names] to [clear-names]
 4. Flatten nested conditionals in [locations]

 Run `/refactor-check-fix [target]` to apply these changes.

═══════════════════════════════════════════

Rules

• DO NOT edit any files
• DO NOT use Edit, Write, or Bash tools that modify files
• Only use Read, Glob, Grep for analysis
• Report findings, suggest fixes, but make no changes
• End with clear next step: /refactor-check-fix