| name | __NAME__ |
| description | Use when … (one-line trigger: name the task, its inputs, and the situation it applies to). This sentence is what the model reads to decide whether to load the skill. |
NAME
Keep this tight — long or reference-heavy detail goes in references/ files the
model opens only when needed (progressive disclosure).
When to use
- The concrete situations this skill handles.
- And the ones it explicitly does NOT — steer the model elsewhere.
How it works
- Step-by-step of what the model does when this skill fires.
- Reference helper files like
references/DETAIL.md instead of inlining everything.
Examples
One worked example: input → action → output.