creating-skills

Create a Skill Unit by initializing it with init_skill.py, which generates a template with YAML frontmatter containing name and description fields. Edit the SKILL.md body to include instructions under 500 lines, organize optional scripts/, references/, and assets/ folders, then validate using verify.py to ensure format compliance.

A SKILL.md file requires YAML frontmatter with name and description, followed by the skill body instructions limited to 500 lines. You may add optional resource folders: scripts/ for automation tools, references/ for documentation, and assets/ for supporting materials, all validated by verify.py.

Run verify.py against your Skill Unit to validate YAML frontmatter, SKILL.md format, and resource folder structure. The script checks that frontmatter contains required fields, body stays under 500 lines, and optional components follow organizational guidelines before packaging.

Yes, this skill supports coaching teams on SKILL.md formatting standards and verify.py implementation. Use the template-driven approach and verify.py validation across all Skill Units to maintain consistent frontmatter, body structure, and resource organization across your skill library.

Creating a new skill uses init_skill.py to generate a fresh template with frontmatter and empty SKILL.md. Updating an existing skill involves editing the current SKILL.md, modifying frontmatter fields, reorganizing resource folders, and re-running verify.py to validate changes while preserving the skill's identity.

This skill depends on quick_validate for validation logic and includes verify.py for format checking. You'll use init_skill.py and package_skill.py as core tools; these handle frontmatter generation, SKILL.md body validation, and resource folder verification without external dependencies.