What is the recommended size for CLAUDE.md?

For optimal performance, CLAUDE.md is ideally under 150 lines. Files exceeding 300 lines are flagged for refactoring into modular rule files.

How does the health score work?

It evaluates your configuration across four categories: size efficiency, logical structure, specificity of rules, and overall maintainability, resulting in a grade from A to D.

Can this skill help me move rules to the .claude/rules folder?

Yes, it identifies specific sections that are over-extended and provides a strategy for migrating them into separate, path-specific rule files.

Why should I audit my CLAUDE.md file?

Large or disorganized files can cause context bloat, making the AI less efficient and potentially leading to slower responses or ignored instructions.

What are the common issues found during a memory audit?

Typical issues include duplicate instructions, abstract rules that lack concrete examples, and 'ghost' information that no longer applies to the current codebase.

Memory Audit

Name: Memory Audit
Author: biwakonbu

bybiwakonbu

Productivity & Workflow

Analyzes and optimizes Claude Code memory configurations to improve project context and AI efficiency.

About

The Memory Audit skill provides a comprehensive framework for evaluating your project's memory setup, specifically focusing on CLAUDE.md and .claude/rules files. It identifies critical optimization points such as file size bloat, redundant instructions, and outdated information, ensuring your AI assistant operates with the most relevant and concise context possible. By utilizing its structured audit reports and health scoring system, developers can maintain a high-performance workspace that minimizes token waste and improves instruction following through modular rule management.

Key Features

Automated size and structure checks for CLAUDE.md and .claude/rules files
0 GitHub stars
Refactoring guidance for migrating bloated files into modular rule sets
Standardized health scoring system (A-D) based on efficiency and maintainability
Detection of redundant, abstract, or outdated technical project instructions
Detailed audit report generation with findings and prioritized action items

Use Cases

Evaluating a large project's CLAUDE.md to prevent context window bloat and instruction drift
Conducting recurring maintenance checks to keep project rules synchronized with the current tech stack
Refactoring monolithic documentation into a modular .claude/rules structure for better organization

About

Key Features

Automated size and structure checks for CLAUDE.md and .claude/rules files
0 GitHub stars
Refactoring guidance for migrating bloated files into modular rule sets
Standardized health scoring system (A-D) based on efficiency and maintainability
Detection of redundant, abstract, or outdated technical project instructions
Detailed audit report generation with findings and prioritized action items

Use Cases

Evaluating a large project's CLAUDE.md to prevent context window bloat and instruction drift
Conducting recurring maintenance checks to keep project rules synchronized with the current tech stack
Refactoring monolithic documentation into a modular .claude/rules structure for better organization