Loading...
Loading...
Author a new Claude Code skill the house way, or refactor a fat skill into a lean SKILL.md + references/. Use when you want to "create a skill", "write a new skill", "turn a prompt or command into a skill", "split a skill into references", "trim a SKILL.md", or invoke /skills-create. The meta-skill — a skill that builds skills.
npx skill4agent add coleam00/skills skills-createreferences/templates/SKILL.mdreferences/SKILL.mddescriptionreferences/skill-standards.md| Type | What it is | Apply |
|---|---|---|
| Workflow | a multi-step procedure (plan, review, ship) | the full lens incl. verifiable validation gates |
| Artifact-generator | produces a document/output | Context-is-King + a suggested (never mandated) output shape |
| Knowledge / reference | facts the agent consults | Context-is-King + information-dense; no phases, no validation loop |
| Tool-wrapper | drives a script / CLI / API | a deterministic script + sharp triggers; validation = the tool's own exit code |
references/skill-standards.md$ARGUMENTS$ARGUMENTScreate [<name>]<name>references/creating-skills.mdrefactor <path/to/SKILL.md>references/refactoring-skills.mdreferences/skill-standards.mdreferences/creating-skills.mdscripts/references/templates/templates/SKILL.template.md.claude/skills/<name>/SKILL.mdreferences/templates/descriptionreferences/validation.mdreferences/refactoring-skills.mdreferences/templates/templates/<x>.mdreferences/validation.mdThe #1 refactor risk: moving an always-needed output format into a lazily-loaded reference, so the agent forgets to read it and the output silently changes. Always pair such an extraction with a mandatory-read line.
references/skill-standards.mdreferences/creating-skills.mdreferences/refactoring-skills.mdreferences/validation.mdtemplates/SKILL.template.md