techdebt

Scan files or folders for technical debt and produce a prioritized triage table.

Updated May 18, 2026
One-click install
npx skills add https://github.com/Noegnesis/noesis --skill techdebt-noegnesis
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: techdebt
Source: https://github.com/Noegnesis/noesis/tree/main/skills/techdebt
Command: npx skills add https://github.com/Noegnesis/noesis --skill techdebt-noegnesis

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

It helps you quickly understand what is technically wrong or risky in a codebase so you can plan cleanup without guessing or starting a refactor blindly.

Core Features & Use Cases

  • Prioritized triage for technical debt: Scans for code smells like long functions, duplication, magic numbers/strings, missing tests, dead code, error swallowing, and coupling.
  • Targeted, scope-limited reading: Supports scanning a specific file or a folder using representative entry points and relevant markers (e.g., TODO/FIXME).
  • Read-only, approval-gated workflow: Outputs a ranked table and waits for you to approve specific findings before any changes are proposed.

Quick Start

Ask for a technical-debt scan by telling the assistant which file or folder you want reviewed, for example: "Run /techdebt on src/ and show the top cleanup priorities."

Frequently Asked Questions about techdebt

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I identify technical debt in a codebase before planning a refactor?

To identify technical debt, run a read-only scan on a specified file or folder to detect code smells, missing tests, magic numbers, untyped boundaries, dead code, and error swallowing. This produces a prioritized triage table to help you plan refactors without guessing.

What is the best way to prioritize code cleanup effort across a module?

The best way to prioritize code cleanup is to scan representative entry points and markers like TODO or FIXME to surface technical debt. The scan outputs a ranked table capped at around 15 items, showing exactly where to invest cleanup effort for the highest impact.

How do I detect code smells and untyped boundaries without auto-fixing my code?

You can detect code smells and untyped boundaries through an approval-gated, read-only scan that avoids auto-fixing. It identifies duplication, dead code, and error swallowing, then waits for you to approve specific findings before any changes are proposed.

Can I scan a specific folder for missing tests and dead code?

Yes, you can scan a specific folder for missing tests and dead code by providing the folder path. The scan uses relevant markers and representative entry points to find technical debt and outputs a prioritized list of issues for your review.

What types of code smells does a technical debt scan look for?

A technical debt scan looks for long functions, duplication, magic numbers and strings, missing tests, dead code, error swallowing, and coupling. It aggregates these code smells into a prioritized triage table to guide your refactoring planning.

Does a technical debt scan automatically modify my source files?

No, a technical debt scan is strictly read-only and does not automatically modify source files. It generates a prioritized triage table of findings and explicitly requests your approval before proposing any specific fixes.