mcpbeat Sign in

Absolute Debt Agent Skill

> Triggers on "absolute debt", "fix our lint warnings", "clear the type errors", "get to strict mode", "burn down suppressions".

3k tokens
context cost
the whole folder, loaded on every use
3
files
instructions only
0
copies elsewhere
how many repositories repackaged it
205
stars on the repo
on the repository, not the skill itself

Install

one command, takes just this skill from the repository
npx skills add https://github.com/maddhruv/absolute --skill absolute-debt

The instruction itself

7 sections, as written by the author

> Start your first response with the 🧾 emoji.

Absolute Debt

Pay down accumulated lint and type-checker debt across the repo — existing warnings,

errors, and suppressions that built up over time — in safe waves, by rule, with tests green

after each. The goal is a genuinely clean lint + typecheck, not a quieter one.

Runs the shared engine in references/health-engine.md — read it for the

DETECT → SCAN → TRIAGE → FIX → VERIFY → REPORT loop and the safety contract. This file

covers only what's specific to lint/type debt.


When to use

  • "Fix our lint warnings", "clean up the type errors", "get the codebase to strict mode".
  • Tightening config (enable a rule, raise tsconfig strictness) and clearing the fallout.
  • Burning down accumulated eslint-disable / @ts-ignore / # type: ignore suppressions.

debt vs simplify: simplify improves *your working diff's* quality. debt clears

pre-existing repo-wide lint/type violations on green main. Use debt for standing

cleanup, not for changes you're actively making.


What it scans

| Ecosystem | Lint debt | Type debt |

|---|---|---|

| JS/TS | eslint . (full report, by rule) | tsc --noEmit |

| Python | ruff check / flake8 | mypy / pyright |

| Go | golangci-lint run | go vet ./..., staticcheck |

Also inventory suppressions as debt in their own right: eslint-disable*,

@ts-ignore/@ts-expect-error, # type: ignore, # noqa, //nolint. Each is a hidden

violation. Count violations grouped by rule (not by file) — that's how you wave them.


Risk ranking (TRIAGE)

| Wave | Class | Default |

|---|---|---|

| 1 | autofixable lint (eslint --fix, ruff --fix, gofmt) | fix now — mechanical, safe |

| 2 | manual lint by rule, low behavioral risk (unused, style, imports) | fix this pass, one rule at a time |

| 3 | type errors, and lint rules that can change behavior if "fixed" wrong | gated — review each, may need real logic |

| 4 | removing suppressions (@ts-ignore etc.) | gated — each may surface a real bug |

Fix one rule across the repo per wave, not one file across all rules — keeps diffs

reviewable and regressions bisectable. Config tightening (enable rule / raise strictness) is

itself a finding: propose it, then clear what it surfaces.


Fix & verify

  • Run the autofixers first (wave 1), commit-ready, then re-scan to see what's truly manual.
  • Manual fixes: address the *cause*, not the symptom. A type error means the types are

wrong somewhere — fix the type, don't cast to any/add # type: ignore.

  • Never clear a violation by suppressing it. Removing suppressions is the *goal*, not a tool.
  • After each wave: full test + lint + typecheck + build. The violation count must drop and

nothing regress.


Gotchas

  • Fixing by suppressing. @ts-ignore to clear a type error *adds* debt. Forbidden here.
  • any / cast to win. Silences the checker, keeps the bug. Fix the real type.
  • One-file-all-rules waves. Mixes concerns, unreviewable. Go one-rule-all-files.
  • Autofix without review. --fix can change behavior (e.g. prefer-const on a

reassigned var via a bug). Run tests after autofix too.

  • Disabling the rule instead of fixing. Loosening config to hit zero is debt laundering,

not paydown.


Companion commands

  • /absolute simplify — quality of code you're *currently changing*.
  • /absolute prune — many unused-var/import warnings vanish once dead code is pruned.
  • /absolute work — if clearing the debt needs real refactoring, hand off.

Other skills for the same job

different authors, same section of the catalogue
Receiving Code Review
by ZhanlinCui
×7

Use when receiving code review feedback, before implementing suggestions, especially if feedback seems unclear or technically questionable - requires technical rigor and verification, not performative agreement or blind implementation

2k tokens
Requesting Code Review
by ZhanlinCui
×6

Use when completing tasks, implementing major features, or before merging to verify work meets requirements

2k tokens
Git Commit
by github
vendor ×3

Execute git commit with conventional commit message analysis, intelligent staging, and message generation. Use when user asks to commit changes, create a git commit, or mentions "/commit". Supports: (1) Auto-detecting type and scope from changes, (2) Generating conventional commit messages from diff, (3) Interactive commit with optional type/scope/description overrides, (4) Intelligent file staging for logical grouping

799 tokens
Github Code Review
by ComeOnOliver
×3

Comprehensive GitHub code review with AI-powered swarm coordination

13k tokens
Karpathy Guidelines
by hyyhf
×3

Behavioral guidelines to reduce common LLM coding mistakes. Use when writing, reviewing, or refactoring code to avoid overcomplication, make surgical changes, surface assumptions, and define verifiable success criteria.

629 tokens
Code Reviewer
by google-gemini
vendor ×2

Use this skill to review code. It supports both local changes (staged or working tree) and remote Pull Requests (by ID or URL). It focuses on correctness, maintainability, and adherence to project standards.

795 tokens
Agent MD Refactor
by softaworks
×2

Refactor bloated AGENTS.md, CLAUDE.md, or similar agent instruction files to follow progressive disclosure principles. Splits monolithic files into organized, linked documentation.

4k tokens
Commit Work
by softaworks
×2

Create high-quality git commits: review/stage intended changes, split into logical commits, and write clear commit messages (including Conventional Commits). Use when the user asks to commit, craft a commit message, stage changes, or split work into multiple commits.

2k tokens

How to use it

Copy the folder

Take maddhruv/absolute-debt from the repository into ~/.claude/skills for personal use, or into .claude/skills inside a project.

Check the name does not clash

The agent identifies a skill by the name field in its header. Two skills with the same name cannot sit side by side — one of them will be ignored.