What problem does it solve? Building custom AI skills requires understanding YAML frontmatter rules, directory layout constraints, and progressive disclosure architecture, and mistakes in any of these cause skills to fail discovery or waste context. ## Core Features & Use Cases - Specification Guidance: Defines the required name and description frontmatter fields with length limits, formatting rules, and valid examples. - Progressive Disclosure Architecture: Explains the 3-level loading model (metadata, SKILL.md body, referenced files) so skills scale without context penalty. - Templates and Checklists: Provides basic, intermediate, and advanced skill templates plus a validation checklist covering structure, content quality, and testing. - Use Case: A developer wants to package a repeatable workflow, such as generating React components, as a discoverable skill for their team, and uses this skill to produce a compliant SKILL.md with scripts and resources directories. ## Quick Start Ask the AI to create a new skill named my-first-skill with a proper SKILL.md including name, description, quick start, and step-by-step instructions.
