claude-md-improver

Audit and improve CLAUDE.md files across a repository.

Updated Apr 1, 2026
One-click install
npx skills add https://github.com/jaimedhenriques/finsyt --skill claude-md-improver-jaimedhenriques
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: claude-md-improver
Source: https://github.com/jaimedhenriques/finsyt/tree/main/artifacts/platform/.agents/skills/claude-md-improver
Command: npx skills add https://github.com/jaimedhenriques/finsyt --skill claude-md-improver-jaimedhenriques

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes scripts (resource) components.

What problem does it solve?

Keeps CLAUDE.md files current, complete, and actionable so Claude Code has reliable project context instead of stale or missing guidance.

Core Features & Use Cases

  • Scans for project-root, package-level, and nested CLAUDE.md variants.
  • Scores each file against coverage, clarity, currency, and actionability criteria.
  • Produces a quality report and prepares minimal, targeted edits after approval.
  • Useful for repository onboarding, memory cleanup, and documentation maintenance across monorepos.

Quick Start

Ask the skill to audit the repository's CLAUDE.md files, summarize the quality gaps, and propose targeted updates.

Frequently Asked Questions about claude-md-improver

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

FAQPage Schema
How do I audit and improve CLAUDE.md files across a monorepo?

To audit CLAUDE.md files across a monorepo, you can scan for project-root, package-level, and nested variants, score them against quality criteria, and apply minimal targeted updates after approval.

What is the best way to maintain project memory and keep CLAUDE.md files current?

Keeping CLAUDE.md files current involves discovery of every variant, quality scoring against coverage, clarity, currency, and actionability rubrics, and producing minimal targeted edits during repository maintenance workflows.

Can I score documentation quality for nested Claude memory files before applying updates?

Yes, you can score nested Claude memory files against a defined rubric to produce a quality report summarizing gaps, ensuring edits are minimal and targeted only after explicit approval.

How do CLAUDE.md files get updated during repository onboarding and cleanup?

During repository onboarding and cleanup, CLAUDE.md files get updated by discovering every variant, scoring quality gaps, and preparing minimal targeted edits to ensure Claude Code has reliable project context.

Does this approach work for package-level documentation maintenance in large codebases?

Yes, this approach works for package-level documentation maintenance by scanning for nested CLAUDE.md variants across large codebases, scoring them against a quality rubric, and proposing targeted updates.