cleanup-audit

Audit repositories for bugs, drift, stale references, and dead artifacts.

15|5|Updated Feb 21, 2026
One-click install
npx skills add https://github.com/sohei56/maul-team --skill cleanup-audit-sohei56
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: cleanup-audit
Source: https://github.com/sohei56/maul-team/tree/main/.claude/skills/cleanup-audit
Command: npx skills add https://github.com/sohei56/maul-team --skill cleanup-audit-sohei56

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve?

This Skill uncovers bugs, stale references, inconsistent documentation, dead artifacts, redundant code, and workflow drift that can accumulate after refactors or before releases.

Core Features & Use Cases

  • Parallel Repository Investigation: Coordinates eight read-only agents across stale references, state consistency, agent and skill consistency, workflow consistency, markdown redundancy, code redundancy, dead hooks, and unused artifacts.
  • Structured Triage: Deduplicates findings, classifies them into five tiers from real bugs to cosmetic issues, and highlights decisions that require user input.
  • Actionable Reporting: Produces per-axis reports, a consolidated synthesis, tier counts, open decisions, and a recommended remediation sequence.
  • Use Case: After a major schema migration or terminology change, run the audit to identify stale documentation, broken references, unregistered hooks, and implementation drift before shipping.

Quick Start

Run the cleanup-audit skill from the maul-team repository to perform a read-only hygiene review and summarize the highest-priority findings.

Frequently Asked Questions about cleanup-audit

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

FAQPage Schema
How do I find stale references and dead code after a major repository refactor?

To find stale references and dead code after a refactor, run a repository audit using parallel multi-agent analysis. This process coordinates read-only agents to inspect your codebase for drift, broken references, and redundant documentation before shipping.

What is the best way to perform a pre-release code hygiene sweep?

The best way to perform a pre-release code hygiene sweep is to run a multi-agent repository audit. It systematically investigates your repository for dead artifacts, unregistered hooks, state inconsistencies, and workflow drift, then triages findings into prioritized tiers.

How does multi-agent analysis work for finding documentation drift?

Multi-agent analysis for documentation drift works by deploying parallel read-only agents across your repository. One agent specifically checks for markdown redundancy and inconsistent documentation, while others inspect code and state contracts, synthesizing results into a consolidated report.

Do I need Claude Code Agent Teams to audit my repository for bugs and workflow inconsistencies?

Yes, you need Claude Code Agent Teams to perform this repository audit. The tool requires agent teams, repository inspection access, git history, and writable temporary storage to execute parallel analysis and synthesize findings into a structured triage report.

What does a repository audit report include when checking for dead artifacts?

A repository audit report for dead artifacts includes per-axis findings, a consolidated synthesis, tier counts, open decisions, and a recommended remediation sequence. It classifies issues from real bugs to cosmetic issues and deduplicates redundant findings.

When should I not use an automated repository audit for maintenance?

You should avoid an automated repository audit if your project lacks accessible git history or writable temporary storage for reports. The multi-agent analysis requires these dependencies to execute parallel investigations and synthesize structured triage results.