agentleFS
LouPineWaysLouPineWays/Loop-Dee-Loup/.claude/skills/skill-maker/SKILL.md

skill-maker

Skill0 starsChanged 3 months agoMITView on GitHub

Create a new repository-local skill when a recurring reasoning or workflow problem has become stable enough to encode, and no existing mechanism already solves it.

21 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

5 rules. 0 are common, 3 are distinctive, and 2 are very specific to this project.

  • 0

    Common

    Most of the 17 edited versions we read keep these.

  • 3

    Distinctive

    This author's own choices. Worth a look.

  • 2

    Very specific

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

  • Common
  • Distinctive
  • Very specific
  1. Very specific:

    Only for work that keeps requiring the same *kind* of judgment across otherwise-different tasks — not a one-off decision, and not something …

    line 8
  2. Distinctive:

    Check for an existing skill, script, governing rule (AGENTS.md, target-repo authority), or persona that already solves it.

    line 13
  3. Distinctive:

    If all three hold, write the skill as:

    line 16
  4. Very specific:

    Keep it as short as sift or model-check — a page, not a manual.

    line 22
  5. Distinctive:

    Do not create a skill for speculative future work, and do not create a "do everything" skill that could be several narrower ones.

    line 24

“Common” is measured: kept by at least 6 in 10 of the 17 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

Skill maker

Only for work that keeps requiring the same kind of judgment across otherwise-different tasks — not a one-off decision, and not something a script could already do deterministically (route that to script-maker).

Before creating anything:

  1. State the concrete recurring problem and point to at least two real instances where it was solved by re-deriving the same reasoning from scratch.
  2. Check for an existing skill, script, governing rule (AGENTS.md, target-repo authority), or persona that already solves it. If one exists, extend or point to it instead of creating a parallel mechanism.
  3. Confirm the problem still requires judgment on each invocation — a fixed recurring role with no per-invocation reasoning is a persona (persona-maker), not a skill.

If all three hold, write the skill as:

  • a narrow trigger: the specific situation that invokes it, phrased so it doesn't fire on unrelated work;
  • the minimum reasoning needed to resolve that situation — no restated project background, no generic advice available elsewhere;
  • an explicit stop condition or output shape, so the skill doesn't sprawl into a general-purpose assistant.

Keep it as short as sift or model-check — a page, not a manual. State how its usefulness will be checked against real work (a specific past or near-term task it should measurably improve), not a hypothetical.

Do not create a skill for speculative future work, and do not create a "do everything" skill that could be several narrower ones.

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.