Copy the block below into domains/<domain>/skills/<skill-name>/skill.md and replace the placeholders. See Authoring a skill for the canonical layout and the full frontmatter schema.
---
name: <skill-name>
description: <one or two sentences, including when-to-use cues so agents can match this skill; keep it well under the operator limits, around 1,000 characters>
maturity: experimental
---
# <Skill Title>
## When To Use
- <the concrete condition that should trigger this skill>
- <another triggering condition>
## Workflow
1. **<First step>.** State what the agent should do and how to tell it is done.
2. **<Second step>.** Keep steps imperative and verifiable.
3. **<Continue as needed>.** Reference supporting material in `references/` rather than inlining long content.
## Common Pitfalls
| Pitfall | Rule |
|---|---|
| <a tempting wrong move> | <the rule that avoids it> |
| <another common mistake> | <its rule> |- Keep
nameunprefixed and matching the skill's directory name. The installer adds themms-prefix to the generated outputs. maturityis optional and defaults tostable. Useexperimentalfor new skills anddeprecatedon the way out.- Put supporting docs in
references/, which the skill reads on demand, rather than expanding the description. The description is always loaded into the agent's discovery surface, while the body and references are not. - Add a
repos/<repo>.mdoverlay only when a consuming repo needs repo-specific guidance. It is merged into the skill body at install time.