| name | skill-name |
| description | [Description] — third person, what the skill does plus a 'Use when...' clause with concrete trigger phrases a user would actually type; 25-60 words; lead with the key use case. |
[Skill Name]
[One or two sentences naming the owned outcome — and, when ownership is
contested, which neighboring skill owns the adjacent work.]
Boundaries
- Use for: [specific triggers and owned outcomes].
- Do not use for: [neighboring responsibilities and their owning skills].
Inputs
- Required: [minimum information needed to begin].
- Optional: [flags, scope, or defaults that change behavior].
- Prerequisites: [tools, credentials, or running state that must exist
before the skill starts — delete this line when there are none].
Workflow
- Confirm the target and constraints.
- Perform the smallest complete workflow that produces the owned outcome.
- Handle explicit failure and ambiguity cases without widening scope.
- Run the verification below; when a check fails, fix the cause and re-run
that check. Repeat until every check passes or a concrete blocker remains,
then report the blocker instead of looping.
Verification
- Run the checks that demonstrate the promised outcome.
- Validate frontmatter with
claude plugin validate --strict <plugin-path>.
- Run repository policy checks with
quick_validate.py <skill-or-plugin-path>.
Completion
Report the outcome, verification performed, changed artifacts, and unresolved
issues. Do not require a fixed report envelope when plain language is clearer.
