| name | my-skill |
| description | Template starter for an Agent Skill — replace this entire description with one that names the concrete intent triggers ("Use when the user asks about X, Y, Z") and lists negative-intent Skip clauses. The platform's skill matcher reads this field — write it for matching, not for marketing. Stay under 1024 chars. |
| user-invocable | false |
My Skill
Template starting point for an Agent Skill (.skill package).
Customize the frontmatter above and the sections below.
See rules/agent-skills.md for authoring rules.
What this skill does
A few paragraphs of conceptual overview. What domain does this cover?
What workflows does it support? What's the user's likely starting
context when activating?
When to use
Be specific. List intent patterns:
- "How do I X" → ...
- "Fix Y" → ...
- File patterns: when editing
**/*.foo → ...
Workflows
Walk Claude through the typical patterns step by step. Keep it actionable.
Workflow 1: ...
- First step
- Second step
Workflow 2: ...
...
Reference
Deep reference details belong in references/X.md (loaded on demand
via Read), not in this top-level SKILL.md. Linking pattern:
Cross-references
If this skill cohabits with siblings, name them and when to defer:
- For X questions: use the
other-skill skill instead
- For Y questions: this skill covers it
