Skill Evaluation and Improvement

This skill focuses on the **evaluation and improvement** of existing skills. In contrast, the `modular-skills` skill focuses on the **design and architecture** for creating new skills. They work together: use this skill to find problems, then use `modular-skills` to implement architectural fixes.

This is a meta-skill that assesses, analyzes, and improves other Claude Code Skills. It runs comprehensive audits against quality standards, analyzes token usage, checks for compliance, and generates prioritized recommendations to enhance skill performance and reliability.

Its core capabilities include a comprehensive skill auditor, an automated improvement suggester, a token usage tracker for efficiency analysis, and a standards and security compliance checker. It provides a detailed evaluation framework with weighted scoring to objectively measure skill quality.

Use this skill for regular audits of your skill library, quality assurance before releasing a new skill, performance tuning to optimize token usage, and ensuring your skills adhere to the latest best practices. It is designed for improving existing skills, not creating new ones from scratch.

It automates the quality assurance process for your skills. Instead of manually checking for issues, you can run audits to get a clear, scored report and a prioritized list of improvements. This saves time, standardizes quality, and helps you maintain a high-performing library of Claude Code Skills.