agentleFS
jared833jared833/claude-code-hooks/skills/make-skill/SKILL.md

make-skill

Skill2 starsChanged 22 days agoMITView on GitHub

Create a new Claude Code skill, either by interviewing me about a job or by capturing a task we just did in this conversation. Use when I say /make-skill, "make a skill", "turn this into a skill", "save this as a skill", or "I keep explaining this".

8 other projects use a version of this.

Adopted by

Most recently taken up by

The file

We split this skill into the rules it gives an agent

8 rules. 0 are common, 8 are distinctive, and 0 are very specific to this project.

  • 0

    Common

    Most of the 8 edited versions we read keep these.

  • 8

    Distinctive

    This author's own choices. Worth a look.

  • 0

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    If we just finished a task in this conversation and I say "turn this into a skill", capture it: pull the steps that actually worked, the …

    line 13
  2. Distinctive:

    Write the description as: what it does, then "Use when I say" plus the trigger phrases, including the plain-English ways I'd ask, not just …

    line 26
  3. Distinctive:

    Write the body: one line on the goal, the steps as a numbered list, the good example, then a "Never" list.

    line 28
  4. Distinctive:

    Keep it under 80 lines. If it needs a template or a checklist, put that in a separate file in the same folder and tell the skill when to …

    line 30
  5. Distinctive:

    If it doesn't fire, the description is wrong.

    line 38
  6. Distinctive:

    Never write a skill for a one-time task.

    line 42
  7. Distinctive:

    Never pad the body with general advice Claude already follows.

    line 43
  8. Distinctive:

    Never overwrite an existing skill without asking.

    line 44

“Common” is measured: kept by at least 6 in 10 of the 8 edited versions of this file we could read. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
SKILL.md

Make a skill

You write skills for me. A skill is one folder with one SKILL.md in it. The description decides when it loads, so most of your effort goes there.

Pick the mode

  • If we just finished a task in this conversation and I say "turn this into a skill", capture it: pull the steps that actually worked, the corrections I made, and the final output.
  • Otherwise, interview me. Ask these one at a time, and wait for each answer:
    1. What's the job, in one sentence?
    2. What would I type or say when I want it? Give me three ways.
    3. What are the steps, in order?
    4. What does a good result look like? Paste one if you have it.
    5. What should it never do?

Write it

  1. Pick a short lowercase name with hyphens. Check ~/.claude/skills/ first, and if the name is taken, ask me before overwriting anything.
  2. Write the description as: what it does, then "Use when I say" plus the trigger phrases, including the plain-English ways I'd ask, not just the slash command.
  3. Write the body: one line on the goal, the steps as a numbered list, the good example, then a "Never" list.
  4. Keep it under 80 lines. If it needs a template or a checklist, put that in a separate file in the same folder and tell the skill when to read it.
  5. Save it to ~/.claude/skills/<name>/SKILL.md.
  6. Show me the finished file and the path you saved it to.

Test it

Tell me to open a fresh session and ask for the job in plain words, without the slash command. If it doesn't fire, the description is wrong. Add the words I actually used and save again.

Never

  • Never write a skill for a one-time task.
  • Never pad the body with general advice Claude already follows.
  • Never overwrite an existing skill without asking.

Download now (ZIP)

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

No reviews yet

People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.

  • It worked
  • Partly
  • It didn't
What did you use it for? e.g. Reviewing pull requests in a Rails app

Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.

Your trail

Files you open will collect here, so you can find them again.

Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.

Now do this for your own.

Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.