manage-claude-md

Manage CLAUDE.md and CLAUDE.local.md lifecycle with tiering and decay-based retirement.

Updated Mar 19, 2026
One-click install
npx skills add https://github.com/sunLeee/optimization --skill manage-claude-md
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: manage-claude-md
Source: https://github.com/sunLeee/optimization/tree/main/.claude/skills/utility/manage-claude-md
Command: npx skills add https://github.com/sunLeee/optimization --skill manage-claude-md

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

CLAUDE.md and CLAUDE.local.md files drift over time, making project context inconsistent. This skill provides structured creation, updating, review, cleanup, and tiering for Claude memory documents to keep a coherent, discoverable knowledge base across a repository.

Core Features & Use Cases

  • Create new CLAUDE.md or CLAUDE.local.md templates tailored to the project.
  • Update and review CLAUDE.md to reflect current tasks, conventions, and architecture.
  • Tier memory content automatically (Level 1/2/3) and perform decay-based retirement when stale.
  • Integrate with memory-workflow and other skills to maintain a crisp, actionable project memory.
  • Use Case: A team runs /manage-claude-md review to generate a structured improvement plan for CLAUDE.md.

Quick Start

Run the command /manage-claude-md review to assess and improve CLAUDE.md quality.

Frequently Asked Questions about manage-claude-md

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

FAQPage Schema
How do I keep project context accurate when CLAUDE.md files drift over time?

To keep project context accurate when CLAUDE.md files drift, you can run a structured review to assess quality, update conventions, and enforce a 3-tier memory model for consistent Claude memory across repositories.

What is the best way to manage stale Claude memory documents in a repository?

The best way to manage stale Claude memory documents is through decay-based retirement, which automatically identifies outdated content and cleans up CLAUDE.md files to sustain a crisp, actionable knowledge base.

How do I structure Claude memory tiering for consistent project documentation?

You structure Claude memory tiering by enforcing a 3-tier memory model that categorizes CLAUDE.md and CLAUDE.local.md content into levels, ensuring project context remains discoverable and concise across repos.

Can I integrate CLAUDE.md cleanup with existing memory-workflow tools?

Yes, you can integrate CLAUDE.md cleanup with memory-workflow tools to maintain long-term reliability, automatically synchronizing project context updates and tiering across your development environment.

When should I create a new CLAUDE.local.md file instead of updating CLAUDE.md?

You should create a new CLAUDE.local.md file for local environment configurations instead of updating CLAUDE.md, keeping shared project context separate from individual repository settings.

Why does inconsistent CLAUDE.md content cause problems for Claude memory management?

Inconsistent CLAUDE.md content causes problems because context drift makes project memory unreliable, requiring structured cleanup and tiering to sustain accurate, actionable documentation across repositories.