harness-maintenance

Enforce repository harness integrity by running task harness:doctor and Biome lint rules.

Updated Jan 19, 2026
One-click install
npx skills add https://github.com/KAFKA2306/yt3 --skill harness-maintenance
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: harness-maintenance
Source: https://github.com/KAFKA2306/yt3/tree/main/.agent/skills/harness-maintenance
Command: npx skills add https://github.com/KAFKA2306/yt3 --skill harness-maintenance

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Enforce and maintain repository infrastructure hygiene to prevent documentation rot, configuration drift, and test decay by ensuring architectural decisions, hooks, linters, and diagnostic checks are created and executed consistently.

Core Features & Use Cases

  • ADR Strict Enforcement: Require creation of ADRs for all architectural decisions, dependency changes, and pattern shifts and store them in docs/adr before proceeding.
  • Hook Integrity: Preserve and restore local hooks in .claude/settings.json and refuse to bypass or weaken hook protections; repair code to satisfy hooks.
  • Doctor Loop & Remediation: Run task harness:doctor at the end of every task or session, surface issues, and require immediate fixes before commit.
  • Linter Primacy & Config Protection: Enforce Biome lint rules (e.g., noExplicitAny: error) and treat biome.json, Taskfile.yml, and .claude/settings.json as immutable unless the harness itself is being evolved.
  • Use Case: Use when making structural changes, dependency upgrades, CI/hook adjustments, or completing task sessions to ensure long-term repository health.

Quick Start

Ask the assistant to validate the repository harness by creating any missing ADRs, running task harness:doctor, and fixing all reported issues.

Frequently Asked Questions about harness-maintenance

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

FAQPage Schema
How do I prevent configuration drift and documentation rot in my repository?

Prevent configuration drift and documentation rot by enforcing repository harness integrity: mandate ADRs for architectural changes, preserve hook configurations, and run diagnostic checks before commits.

How do I enforce ADR creation for architectural decisions and dependency upgrades?

Enforce ADR creation by requiring architectural decisions, dependency changes, and pattern shifts to be documented in the docs/adr directory before any structural code changes can proceed.

Can I bypass hook protections in .claude/settings.json to speed up commits?

No, hook protections in .claude/settings.json must be preserved and restored. The harness refuses to bypass or weaken hook protections; instead, repair the code to satisfy the hook requirements.

What is the best way to validate repository health at the end of a development session?

The best way to validate repository health is executing task harness:doctor at the end of every session, surfacing all reported issues, and requiring immediate remediation before any code commit.

Does this repository health enforcement tool work with Biome linter rules?

Yes, it enforces Biome lint rules such as noExplicitAny: error and protects biome.json, Taskfile.yml, and .claude/settings.json as immutable configurations unless the harness itself is being evolved.

When should I not modify biome.json or Taskfile.yml in my project?

You should not modify biome.json, Taskfile.yml, or .claude/settings.json unless you are intentionally evolving the repository harness itself, as these files are treated as immutable configuration protections.