tech-debt

name	tech-debt
description	Scan codebase for technical debt and create GitHub Issues for high-severity findings
user-invocable	true
disable-model-invocation	true
allowed-tools	["Glob","Grep","Read","Agent","Bash(gh issue create:*)","Bash(gh issue list:*)","Bash(git log:*)","TaskCreate","TaskUpdate","TaskList","AskUserQuestion"]

/tech-debt — Technical Debt Scanner

Scan the codebase for technical debt and refactoring opportunities, then create GitHub Issues for actionable items.

Steps

1. Scan the codebase for the patterns listed below
2. For each finding, assess severity (High / Medium / Low) and effort (S / M / L)
3. Group findings by category
4. For High severity items, create GitHub Issues using gh issue create
5. Output a summary report

Detection Patterns

Code Smells

• Duplicated code (3+ similar blocks)
• Long functions (50+ lines)
• Large files (500+ lines)
• Deep nesting (4+ levels)
• God classes (too many responsibilities)
• Dead code (unused functions, imports, variables)
• TODO/FIXME/HACK comments left in code

Architecture

• Circular dependencies between modules
• Layer violations (check CLAUDE.md for architecture boundaries)
• Missing abstractions (concrete classes where interfaces should be)
• Inconsistent patterns across similar features

Performance

• N+1 queries
• Missing database indexes on frequently queried columns
• Unbounded lists without pagination
• Heavy computation on main thread
• Memory leaks (retain cycles, leaked references)

Testing

• Public functions without unit tests
• Test files with no assertions
• Missing edge case coverage (null, empty, boundary values)

Dependencies

• Outdated dependencies (check against latest stable versions)
• Deprecated API usage
• Security vulnerabilities in dependencies (if audit tools available)

Issue Creation

For each High severity finding:

gh issue create \
  --title "Tech Debt: {concise description}" \
  --body "{detailed description with file locations and suggested fix}" \
  --label "tech-debt"

Output Format

## Tech Debt Report

**Scanned:** N files
**Findings:** X High, Y Medium, Z Low
**Issues created:** N

### High Severity
| # | Category | File | Description | Effort | Issue |
|---|----------|------|-------------|--------|-------|

### Medium Severity
| # | Category | File | Description | Effort |
|---|----------|------|-------------|--------|

### Low Severity
(summary only)