Guides the architecture and scaffolding of high-quality agent skills following the open agent-skills standard.
This skill provides a comprehensive framework for developers to create, organize, and validate specialized capabilities for Claude Code. It enforces the open agent-skills standard through a structured workflow that includes defining clear triggers, implementing progressive disclosure via hierarchical documentation, and avoiding common anti-patterns like information overload. By using this guide, developers can ensure their custom skills are concise, modular, and optimized for maximum agent performance across different AI models.
주요 기능
01Generate optimized SKILL.md files with YAML frontmatter
02Validate skill readiness using comprehensive quality checklists
03Apply best practices for task-specific degrees of freedom
04Scaffold standardized directory structures for new agent skills
05Implement progressive disclosure to manage agent context efficiently
060 GitHub stars
사용 사례
01Building a new library of domain-specific capabilities for Claude Code
02Optimizing agent performance by migrating heavy documentation to the references/ pattern
03Standardizing existing internal documentation into agent-executable skills
