agentleFS
mypip-iomypip-io/skills/expert-to-skill/SKILL.md

expert-to-skill

Skill1 starsChanged 25 days agoMITView on GitHub

Turns an expert's or thought leader's public posts, newsletters, talks and transcripts into a reusable Claude skill (a SKILL.md file) that applies their method, with credit and without copying their work. Also works on a founder's own interviews and notes. Use when someone wants to create a Claude skill, build a skill from someone's content, make a skill from their own expertise, or asks how to write a SKILL.md.

The file

We split this skill into the rules it gives an agent

5 rules. About 0 are common, 4 are distinctive, and 1 is very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈4

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 1

    Very specific

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

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    Test it: Run the skill on two real situations.

    line 35
  2. Distinctive:

    Don't put words in their mouth.

    line 52
  3. Distinctive:

    Keep it practical: steps someone can follow today, not a biography.

    line 53
  4. Distinctive:

    Don't publish a skill that presents itself as the expert's official work unless they agree.

    line 54
  5. Very specific:

    When the MyPip MCP is live, skills you build this way will be able to run on the people and replies in your MyPip account, so the method …

    line 79

≈ “Common” is an estimate: at least 1 in 5 of the 350,495 skill files we read have a rule of the same kind, not the same words. “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

Expert to skill

Many experts already publish their whole method for free, one post at a time. This skill collects it into something you can use.

Thought leaders post their frameworks in public every week: a rule in one post, a story in another, a worked example in a third. Read together, those posts often describe a complete way of working. This skill reads a set of someone's public content, pulls out the method (their steps, rules, tests and examples), and writes it as a Claude skill that applies that method to your situation, with credit to them.

It works just as well on your own material: call transcripts, voice notes, interview answers, old docs. Systems that work from your own words beat AI that writes the average version of what everyone says.

What you need

  1. The source material: pasted posts, newsletter issues, transcripts, or links if you can browse. Ten or more pieces is a good start.
  2. Who the skill is for and what job it should do. "Help a founder write a first message the way this person teaches" is a job. "Be like this person" is not.
  3. The expert's name and where their work lives, for credit.

Steps

  1. Read everything once, without summarising. Note each distinct idea, rule, test, story and example, with the piece it came from.
  2. Find the method. Group the notes into: what the expert believes (the principles), what they tell you to do (steps), how they check work (tests), what they warn against (anti-patterns), and the examples they use. Discard anything that's only opinion with no instruction.
  3. Resolve contradictions. Experts change their minds. Where two posts disagree, prefer the newer one and note it.
  4. Pick the job. A skill does one job well. If the material covers several jobs, write several skills.
  5. Write the SKILL.md.
    • Frontmatter: name (lowercase, hyphens) and description, which says what the skill does and exactly when to use it, including the phrases someone would type. The description decides when the skill is used, so make it specific.
    • Body: a short explanation of the idea, what the skill needs, numbered steps, the output format, rules, and one worked example.
    • Credit: name the expert and link their work. Summarise their method in your own words.
  6. Test it. Run the skill on two real situations. Check the output sounds like the method, not like generic advice. Tighten the steps where it drifted.

Output

A complete SKILL.md, plus:

Sources used: <list with links>
Ideas included: <list>
Ideas left out, and why: <list>
Test results: <what changed after testing>

Rules

  • Credit the expert clearly, and link to their work.
  • Paraphrase. Never paste long passages, paid course material, or anything behind a paywall.
  • Don't put words in their mouth. If the method has a gap, say the skill fills it and that the expert didn't say it.
  • Keep it practical: steps someone can follow today, not a biography.
  • Don't publish a skill that presents itself as the expert's official work unless they agree.

Skill format reference

---
name: my-skill-name
description: What it does, and when to use it, including the phrases people would type.
---

# Title

Why this works, in a short paragraph.

## What you need
## Steps
## Output
## Rules
## Example

Save it as SKILL.md in a folder with the same name as the skill. In Claude Code, put the folder in ~/.claude/skills/ (or .claude/skills/ inside a project). In the Claude app, add it as a custom skill in settings.

With MyPip (coming)

When the MyPip MCP is live, skills you build this way will be able to run on the people and replies in your MyPip account, so the method gets applied to real people every day.

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.