tech-debt_skill

Overview

This skill helps identify, prioritize, and remove technical debt across a codebase. It focuses on finding code smells, TODOs, dead code, unused imports, and type issues, then categorizing and fixing them safely. The goal is less code, fewer bugs, and faster maintenance cycles.

How this skill works

The skill scans the repository for debt indicators (TODO/FIXME, long functions, unused imports, bare excepts, Any/type: ignore, duplicated blocks). It classifies each finding by severity, effort, and risk, then proposes a prioritized remediation plan separating quick wins from complex refactors. When authorized, it can apply atomic fixes, run tests, and document changes in a concise tech-debt report.

When to use it

• Before major feature work to reduce risk from hidden debt
• During scheduled code health passes or sprint clean-up days
• When tests or CI are failing due to type or import issues
• If the codebase contains many TODO/FIXME comments
• When onboarding and trying to simplify complex modules

Best practices

• Run tests after each atomic change and keep commits focused
• Prioritize low-effort, high-impact quick wins first
• Avoid mixing debt fixes with new features in the same change
• Check for dynamic usages (getattr, eval, reflection) before deletion
• Document complex fixes with a plan and link to an issue tracker

Example use cases

• Remove unused imports and variables across Python modules and verify types
• Replace bare except blocks with specific exceptions and add fallback behavior
• Detect duplicated handlers and propose abstraction to consolidate logic
• Scan for TODO/FIXME comments and convert high-priority items into tracked issues
• Produce a Tech Debt Analysis report listing critical, quick-win, and planned items