How does the health score work?

It provides a 0-100 rating where 90+ is excellent. It weighs security at 100%, official compliance at 80%, and general best practices at 70%.

Can I use this in my GitHub Actions or CI/CD pipeline?

Yes, the skill includes a JSON reporting mode designed specifically for machine-readability and automated quality gate integration.

Can this skill detect security vulnerabilities?

Yes, it specifically checks for CRITICAL security risks including API keys, tokens, passwords, and private connection strings within your configuration.

What is the primary purpose of the CLAUDE.md Auditor?

It validates project memory files (CLAUDE.md) against official standards and best practices to ensure they are secure, lean, and optimized for Claude's performance.

CLAUDE.md Auditor

Name: CLAUDE.md Auditor
Author: cskiro

bycskiro

•

Security & Testing

Audits and optimizes CLAUDE.md configuration files to ensure security compliance, adherence to official Anthropic guidance, and peak LLM context performance.

The claude-md-auditor is a specialized tool designed to validate and score CLAUDE.md files, which serve as the primary memory and instruction layer for Claude Code. It evaluates configurations against three authoritative sources: official Anthropic requirements, community best practices, and academic research on LLM attention patterns. This skill is essential for developers who want to prevent configuration drift, eliminate security risks like hardcoded secrets, and ensure their project context remains lean and effective for AI-driven development.

Key Features

01JSON output mode for seamless integration into CI/CD automated quality gates

02Automated refactoring to generate optimized, production-ready CLAUDE.md files

03Health scoring (0-100) across security, compliance, and optimization categories

042 GitHub stars

05Validation against official Anthropic memory hierarchy and import syntax

06Comprehensive security scanning for secrets, API keys, and private tokens

Use Cases

01Automating quality control within CI pipelines to maintain documentation standards

02Onboarding new repositories to establish standardized AI interaction guidelines

03Pre-commit auditing of memory file changes to prevent context bloat or security leaks

What are Skills?·How to Install

Install with 🐟 Skill.Fish

npx skillfish add cskiro/claudex claude-md-auditor

For use in Claude.ai and ChatGPT

Download Skill