skill-creator

To create an AI skill, structure it with a SKILL.md file alongside bundled resources like scripts, references, and assets. Follow best practices for conciseness, progressive disclosure, and iterate through initialization, editing, and packaging.

Progressive disclosure in AI skill development is a design principle that structures SKILL.md and bundled resources to reveal complexity gradually. It ensures AI execution receives appropriate degrees of freedom without being overwhelmed by unnecessary details.

The best way to write a SKILL.md file is to keep it concise and focused. Outline specialized workflows, set appropriate degrees of freedom for AI execution, and use bundled references to handle detailed documentation separately.

Yes, understanding prompt engineering is essential. You need to define appropriate degrees of freedom for AI execution and apply design principles like conciseness to ensure the modular skill functions correctly.

To package and iterate on a newly created skill, follow the step-by-step creation process: understand examples, edit the SKILL.md and bundled components, package the modular structure, and continuously iterate based on execution feedback and workflow design needs.

You should bundle scripts and references into your AI skill when the core task requires specialized tools or detailed documentation. This modular approach extends AI capabilities while keeping the main SKILL.md concise through progressive disclosure.