---
# Lowercase, hyphens only. This becomes the folder name too:
#   .claude/skills/<name>/SKILL.md
name: your-skill-name

# This line decides whether the skill ever fires. Claude reads it to decide
# when to load the skill, so write it for the trigger, not for a human reader.
#   - Say what the skill does in plain words.
#   - List the actual phrases you'd type: "Use whenever I ask to ..., ..., or ...".
# "Helps with git." never fires. "Commit my changes the way I do it. Use whenever
# I ask to commit, write a commit message, or summarize my changes." does.
description: Do <the thing> the way I do it. Use whenever I ask to <phrase>, <phrase>, or <phrase>.
---

# How I <do the thing>

<!--
  Write the steps you'd give a new hire on their first day.
  Numbered, in order, specific to YOUR setup. If a step depends on a command,
  name the command. If it depends on a convention, spell the convention out.
-->

1. <First thing to check or run. Say what to do if it fails.>
2. <Where to get the info it needs, e.g. ticket number from the branch name.>
3. <The format the output should take. Be concrete: lengths, order, wording.>

<!--
  Hard rules. The stuff you're tired of fixing by hand.
  Phrase them as "Never ..." or "Always ..." so there's no wiggle room.
-->

4. Never <the output you keep having to correct>.

<!--
  Before you call it done:
  - Ask Claude to do the task with words from your description. Did it fire?
  - If not, the description is the problem, not the steps. Add the phrases you used.
  - Keep it short. One skill, one job.
-->