Skill Invocation - Claude Code
Trigger with /skillname or let Claude auto-load when relevant.
Skills can be invoked two ways: explicitly with a slash command, or automatically when Claude recognizes a matching task.
What it does
Type /skillname and Claude runs that skill with its predefined instructions. Or just describe a task in natural language - Claude reads the available skill descriptions and picks one if it matches. The auto-invocation path is why skill frontmatter quality matters so much.
When to use it
- Explicit invocation when you know exactly which skill you want.
- Auto-invocation when you want Claude to pick the right pattern unprompted.
- Mixing both: auto for discovery, explicit for reliability.
- Teaching new teammates - show them the slash command, let auto-invocation do the rest.
Gotchas
- Auto-invocation sometimes picks the wrong skill if descriptions overlap. Be surgical with descriptions.
- Explicit invocation overrides any safety checks in the description heuristics.
- Skills triggered by auto-invocation still respect permission rules and hooks.
Official docs: https://code.claude.com/docs/en/skills.md
Technical content at the intersection of AI and development. Building with AI agents, Claude Code, and modern dev tools - then showing you exactly how it works.
Get the weekly deep dive
Tutorials on Claude Code, AI agents, and dev tools - delivered free every week.
Was this helpful?
