claude-md-audit

Audit codebases and generate CLAUDE.md documentation with parallel Explore agents.

Updated Mar 4, 2026
One-click install
npx skills add https://github.com/stitts-dev/dotfiles --skill claude-md-audit-stitts-dev
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: claude-md-audit
Source: https://github.com/stitts-dev/dotfiles/tree/main/.claude/skills/claude-md-audit
Command: npx skills add https://github.com/stitts-dev/dotfiles --skill claude-md-audit-stitts-dev

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes scripts (resource) and references (resource) components.

What problem does it solve?

This Skill automates the comprehensive auditing and generation of CLAUDE.md documentation by leveraging parallel AI agents to deeply analyze codebases.

Core Features & Use Cases

  • Parallel Analysis: Employs three distinct "Explore" agents (WHY, WHAT, HOW) simultaneously for thorough codebase examination.
  • Quality Scoring: Assesses documentation completeness and accuracy against codebase reality using a defined scoring framework.
  • Content Generation: Produces ready-to-use documentation sections based on agent findings, filling gaps and updating stale content.
  • Use Case: When onboarding to a new project, use this skill to instantly generate a foundational CLAUDE.md that accurately reflects the project's purpose, architecture, and development workflows.

Quick Start

Run the claude-md-audit skill to analyze the current codebase and generate documentation.

Frequently Asked Questions about claude-md-audit

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

FAQPage Schema
How do I generate CLAUDE.md documentation for a new codebase?

CLAUDE.md documentation is generated by launching three parallel Explore agents to analyze codebase purpose, architecture, and workflows, synthesizing findings into ready-to-use content. It instantly produces a foundational documentation file for project onboarding.

What is the best way to audit existing CLAUDE.md files against codebase reality?

The best way to audit CLAUDE.md files is by using a defined scoring framework to assess documentation completeness and accuracy against codebase reality. This identifies gaps and stale content by comparing documentation to the project's current state.

How does the parallel agent analysis work for codebase documentation?

Parallel agent analysis deploys three distinct Explore agents simultaneously to examine the WHY, WHAT, and HOW of a codebase. This concurrent execution thoroughly maps project purpose, architecture, and development workflows for documentation synthesis.

Can I use this documentation audit on an existing project with stale content?

Yes, this audit works on existing projects with stale content. It assesses documentation completeness and generates updated sections based on agent findings, filling gaps and replacing outdated information to accurately reflect the current codebase.

Do I need any dependencies to run the codebase analysis?

No external dependencies are required to run the codebase analysis. The skill operates independently using its internal scripts and references to launch the parallel Explore agents and synthesize the CLAUDE.md documentation content.