cc-refresh

Audit CLAUDE.md files for accuracy and quality against six criteria.

Updated Mar 17, 2026
One-click install
npx skills add https://github.com/SPRAGE/dev-template --skill cc-refresh
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: cc-refresh
Source: https://github.com/SPRAGE/dev-template/tree/main/template/.claude/skills/cc-refresh
Command: npx skills add https://github.com/SPRAGE/dev-template --skill cc-refresh

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Audit Claude Code context including CLAUDE.md, knowledge store, auto-memory files, session history, and guardrails, identify stale content, and drive targeted maintenance.

Core Features & Use Cases

  • Audits CLAUDE.md files for accuracy and quality, scoring against a defined rubric.
  • Audits the knowledge store, memory, and rules for consistency with the codebase.
  • Generates actionable changes and reports, with optional dry-run to preview updates before applying.

Quick Start

Run the cc-refresh skill to audit CLAUDE.md, memory, and rules and apply approved updates.

Frequently Asked Questions about cc-refresh

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

FAQPage Schema
How do I audit CLAUDE.md files for accuracy and stale content?

To audit CLAUDE.md files for stale content, run an automated context audit that scores commands, architecture clarity, non-obvious patterns, conciseness, currency, and actionability, returning structured findings and a final grade with any red flags.

What is a staleness audit for Claude Code context?

A staleness audit for Claude Code context evaluates CLAUDE.md files, knowledge stores, auto-memory files, and guardrails against the current codebase to identify outdated workflows, architecture inconsistencies, and deprecated conventions.

Can I preview updates before applying changes to my CLAUDE.md knowledge store?

You can preview updates before applying changes to your CLAUDE.md knowledge store by using the optional dry-run feature, which generates actionable changes and reports for review without modifying the actual files.

What criteria are used to score CLAUDE.md file quality?

CLAUDE.md file quality is scored against six criteria: Commands and workflows, Architecture clarity, Non-obvious patterns, Conciseness, Currency, and Actionability, ensuring context accurately reflects the current stack and conventions.

Does the Claude Code context audit check memory and rules for consistency?

The Claude Code context audit checks memory, rules, and the knowledge store for consistency with the codebase, identifying stale content and driving targeted maintenance to keep guardrails aligned with active development.