claude-memory-curator

Audit and clean Claude Code durable context files and settings.

4|1|Updated May 19, 2026
One-click install
npx skills add https://github.com/stark-ai-de/agent-skills --skill claude-memory-curator
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: claude-memory-curator
Source: https://github.com/stark-ai-de/agent-skills/tree/main/skills/claude-operations/claude-memory-curator
Command: npx skills add https://github.com/stark-ai-de/agent-skills --skill claude-memory-curator

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill requires node, and includes scripts (resource) and references (resource) and assets (resource) components.

What problem does it solve?

This Skill aids in the audit, review, and cleanup of Claude Code durable context, ensuring it remains accurate, secure, and effective.

Core Features & Use Cases

  • Context Audit: Inspect and evaluate the integrity of Claude Code memory and instructions.
  • Cleanup & Pruning: Remove stale, conflicting, or unsafe entries from Claude Code durable context.
  • Risk Management: Identify and mitigate risks associated with Claude Code memory pollution and sensitive context.
  • Use Case: When you need to review Claude Code memory for potential issues, or when you want to ensure that sensitive information is securely managed.

Quick Start

Run the claude-memory-curator skill with the --repo flag to specify the repository you want to audit.

Frequently Asked Questions about claude-memory-curator

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

FAQPage Schema
How do I audit Claude Code durable context for security and integrity?

You can audit Claude Code durable context by running the curator skill with the --repo flag to inspect memory, rules, and settings for risks, conflicts, and sensitive information.

What is the best way to clean up stale entries in Claude Code memory?

To clean up Claude Code memory, the curator skill performs pruning operations that identify and remove stale, conflicting, or unsafe entries from your durable context files.

Do I need Node.js to run Claude Code context cleanup operations?

Yes, Node.js is required to run the curator skill, as it depends on specific Node packages for scanning memory entries and editing Claude Code durable context files.

How does risk assessment work for Claude Code memory pollution?

Risk assessment for Claude Code memory pollution involves inventorying your context files and settings to identify sensitive information and mitigate potential security risks.

Can I review Claude Code rules and settings for conflicting instructions?

Yes, the curator skill inspects and evaluates the integrity of Claude Code rules and settings, allowing you to review and resolve any conflicting or unsafe instructions within your repository.

When should I not use automated context cleanup on my Claude Code files?

You should be cautious with automated context cleanup when dealing with critical durable context entries, as the pruning process removes stale or conflicting data that may still be needed.