Skills DirectorySkills Directory
SkillsLearnSecurityCategoriesDocsBlogPro
Sign InSubmit Skill
Skills Directory

Security-tested agent skills for Claude, coding agents, and AI workflows.

Directory

  • Browse Skills
  • All Skills A–Z
  • Claude Skills
  • Claude Code Skills
  • Agent Skills
  • Categories
  • Authors
  • Submit a Skill

Learn

  • Learn Hub
  • Install Claude Skills
  • Write SKILL.md
  • Skills vs MCP
  • Directories Compared

Security

  • Security
  • Methodology
  • Secure Claude Skills
  • Security Badges
  • Chrome Extension
  • Skill Manager

Company

  • About
  • Community
  • Blog
  • API Docs
  • Advertise

2026 Skills Directory. All rights reserved.

ProTermsPrivacyRefunds
Back to skills

Tech Debt

ASecurity

Document and prioritize technical debt. Use when the user says "tech debt", "legacy code", "this code is a mess", "refactoring plan", "we need to clean this up", "paying down debt", "this is hard to change", "how do we fix this codebase", or has code that's difficult to understand or modify - even if they don't explicitly say "technical debt".

20 stars
0 votes
0 copies
0 views
Added 10/4/2026
ai-agentsgorefactoring

Security Analysis

A100/100

Scanned 10/4/2026

$npx -y skills add qa-aman/claude-skills --skill tech-debt --agent claude-code

Installs into .claude/skills of the current project.

Are you the author of Tech Debt?

Add the live security badge to your README — it updates automatically with every re-scan.

Security grade badge for Tech Debt
[![Security: A — Skills Directory](https://www.skillsdirectory.com/api/skills/qa-aman-tech-debt/badge)](https://www.skillsdirectory.com/skills/qa-aman-tech-debt)

More formats (shields.io, HTML) on the badges page. Keep it an A: scan every change in CI with Pro.

Download with Pro
Files
SKILL.md
---
name: tech-debt
description: >
  Document and prioritize technical debt. Use when the user says "tech debt",
  "legacy code", "this code is a mess", "refactoring plan", "we need to clean this up",
  "paying down debt", "this is hard to change", "how do we fix this codebase",
  or has code that's difficult to understand or modify
  - even if they don't explicitly say "technical debt".
---

## Overview

Based on **"Working Effectively with Legacy Code"** by Michael Feathers. Feathers' definition: legacy code is code without tests. His central technique: find the "seams" - places where you can insert test coverage without changing the code under test. Once a seam has test coverage, you can safely refactor. The discipline: never refactor without a safety net. Never add features to untested code without first covering it.

## Workflow

### Step 1: Define the debt
Before prioritizing, describe the debt specifically:

```
Debt item: [name]
Location: [file:line or module]
What it is: [one sentence description]
Why it's a problem: [what it makes harder]
Symptom: [what developers experience because of it]
```

Vague debt ("the codebase is messy") can't be prioritized. Specific debt can.

### Step 2: Classify the debt type
| Type | Description | Urgency |
|------|-------------|---------|
| **Reckless/Deliberate** | Shortcuts taken knowingly ("we'll fix it later") | High |
| **Reckless/Inadvertent** | Mistakes made without knowing better | High |
| **Prudent/Deliberate** | Conscious tradeoff: ship now, fix later | Medium |
| **Prudent/Inadvertent** | Learned better approach after the fact | Low |

### Step 3: Find the seams (Feathers)
A seam is a place where you can change behavior without editing existing code - typically a function call boundary or an interface.

For each debt item:
- Where is the seam that would let you add tests without changing production code?
- What would you need to mock or stub to isolate this code?
- What's the minimum test coverage needed before safe refactoring?

### Step 4: Estimate the cost of the debt
| Dimension | Questions |
|-----------|-----------|
| **Cost of carry** | How much does this slow development each sprint? |
| **Cost of fix** | How many days to resolve? |
| **Blast radius** | How many systems/teams are affected? |
| **Risk if ignored** | What breaks if this is never fixed? |

### Step 5: Prioritize the backlog
Prioritize debt that:
- Has high cost of carry (slows the team every sprint)
- Sits in high-change areas (touched frequently)
- Blocks a planned feature

Deprioritize debt in stable areas that are rarely touched.

### Step 6: Write the debt ticket
```
Title: [specific debt item]
Location: [file/module]
Type: [reckless/prudent x deliberate/inadvertent]
Impact: [what this costs the team per sprint]
Fix plan:
  1. Add seam tests: [specific tests to write first]
  2. Refactor: [what to change once tests exist]
  3. Verify: [how to confirm the refactor is safe]
Effort estimate: [days]
Priority: [P1/P2/P3]
```

## Anti-Patterns

**1. Refactoring without tests**
Bad: "Let me clean this up" without adding test coverage first.
Good: Find the seam. Write tests. Then refactor. Tests prove you didn't break behavior.

**2. Big bang rewrites**
Bad: "We'll rewrite the whole module from scratch."
Good: Feathers: big rewrites fail. Incrementally improve with seam-by-seam coverage. Ship working code at every step.

**3. Paying debt in high-risk sprints**
Bad: Refactoring core payment code the week before a major release.
Good: Pay debt in low-risk windows. Prioritize timing as much as priority.

**4. Debt without business context**
Bad: "This code is bad, we need to fix it." (to a PM or stakeholder)
Good: "This debt costs us 2 days per sprint. Fixing it is a 5-day investment that pays back in 3 sprints."

## Quality Checklist

- [ ] Debt is described specifically (file, module, or function - not "the codebase")
- [ ] Debt type classified
- [ ] Seam identified for each item (where to add tests safely)
- [ ] Cost of carry estimated in developer-days per sprint
- [ ] Cost to fix estimated
- [ ] Highest priority items are in high-change areas
- [ ] Debt ticket includes fix plan with seam tests as first step

Attribution

qa-amanqa-aman
View sourceSee grades on GitHubMore from qa-aman →
SSkills DirectorySkills Directory

Ship a skill? Prove it's safe.

Free 120-pattern security scan, letter grade, and an embeddable README badge.

Submit a skill

Is this your skill, or is something wrong with this listing? Request removal or report an issue. Author removals are honored within 72 hours.

Comments (0)

No comments yet. Be the first to comment!

SSkills DirectorySkills Directory

Ship a skill? Prove it's safe.

Free 120-pattern security scan, letter grade, and an embeddable README badge.

Submit a skill

Related Skills

Caveman

Terse caveman voice: answer first, fluff gone, every technical fact kept. Use for /caveman, "caveman mode", "talk like caveman", "be brief", "less tokens". Stays on until "stop caveman" or "normal mode".

1100021 votes

Hyperplan

Adversarial multi-agent planning skill. Self-orchestrates 5 hostile category members (unspecified-low, unspecified-high, deep, ultrabrain, artistry) via team-mode for ruthless cross-critique debate, distills only the defensible insights, then MANDATORILY hands the distilled insight bundle to the `plan` agent for executable plan formalization. Use when planning needs maximum rigor and surfacing of weak assumptions, blind spots, and over-engineering. Triggers: 'hyperplan', 'hpp', '/hyperplan', ...

698621 votes

Writing Skills

Create and manage Claude Code skills in HASH repository following Anthropic best practices. Use when creating new skills, modifying skill-rules.json, understanding trigger patterns, working with hooks, debugging skill activation, or implementing progressive disclosure. Covers skill structure, YAML frontmatter, trigger types (keywords, intent patterns), UserPromptSubmit hook, and the 500-line rule. Includes validation and debugging with SKILL_DEBUG. Examples include rust-error-stack, cargo-dep...

3931 votes

Mcp Code Execution

Routes multi-tool workflows through MCP servers for large datasets and pipelines. Use when Bash tool overhead is limiting throughput on data-heavy tasks.

3421 votes

catchup

Recovers the conversation and failed tool calls of a previous Codex, Amp, Claude Code, Antigravity, Cline, Copilot CLI, Cursor, DeepSeek Harness, Grok Build, Kimi, OpenCode, Pi Agent, or ZCode session. Use when the user says "catch up", "what did the last session do", "get me up to speed", "I switched agents", asks to recover/summarize a previous session before continuing, or asks to diagnose or report a catchup failure. Do NOT use for the current conversation, git history, or any non-agent log.

741 votes
View all in ai-agents →