skill-creator

To create an AI skill, you structure a SKILL.md file with specialized knowledge, bundle necessary scripts and references, and follow progressive disclosure principles to manage agent context efficiently. This ensures modular, self-contained extensions.

Progressive disclosure in AI skill development is managing context efficiently by layering information across metadata, the SKILL.md body, and bundled resources like scripts and references. This prevents overwhelming the agent while providing necessary workflows.

Structure AI skill resources by placing executable code in a scripts directory, documentation in a references directory, and other static files in an assets directory. This organization maintains a modular, self-contained extension for AI agents.

Yes, you can update existing AI skills by following the structured guidance for SKILL.md content and resource bundling. The workflow facilitates iterating on and packaging modular extensions to refine agent capabilities.

The best way to document internal APIs for an AI agent is to organize the API documentation within a references directory and outline the interaction workflows in the SKILL.md file, ensuring the skill is self-contained and modular.