Provides authoritative guidance and best practices for creating, reviewing, and optimizing Claude Code skill files.
This skill streamlines the development of Claude Code extensions by providing a standardized framework for authoring SKILL.md files. It offers real-time guidance on naming conventions, frontmatter configuration, and structural patterns like input validation and progress checklists, ensuring that custom skills are concise, discoverable, and performant. Whether you are drafting a new skill from scratch or troubleshooting an existing one, this utility ensures your agent capabilities follow Anthropic's best practices for progressive disclosure and token efficiency.
主要功能
01Automated SKILL.md naming and frontmatter validation
02Structural patterns for error recovery and multi-step checklists
03Best practices for concise, third-person skill descriptions
040 GitHub stars
05Support for dynamic context injection and output templates
06Implementation of progressive disclosure for token efficiency
使用场景
01Standardizing documentation and configuration across an agent marketplace
02Drafting new custom Claude Code skills from scratch
03Refactoring existing skill files to improve discovery and performance
