agentleFS
Sign inStart free
nuttakulsvnuttakulsv/think-like/skills/mind-maker/SKILL.md

mind-maker

Skill2 starsChanged 41 days agoMITView on GitHub

Build a new mind skill for this pack from sources the user provides (books, papers, talks, interviews, blog posts, a mentor's own notes): extract the person's methods, heuristics and anti-patterns into a SKILL.md with the same structure as the other minds, with sources and an ethics check for living people. Use when the user wants to add a person, mentor, or expert to think-like, or turn someone's way of working into an agent skill. Triggers: "make a mind", "create a skill from this person", "add <name> to think-like", "distill how X thinks", "/mind-maker", "สร้าง skill จากคนนี้", "ทำ skill ของ", "เพิ่มอัจฉริยะ".

The file

We split this skill into the rules it gives an agent

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

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈3

    Distinctive · estimate

    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. Distinctive:

    Ask for, or locate, 2–5 primary sources: the person's own books, papers, talks, letters, posts.

    line 29
  2. Very specific:

    Create skills/<slug>/SKILL.md with this structure (match the other minds):

    line 46
  3. Distinctive:

    Keep it 60–120 lines. Method steps must be things an agent can do with code, docs, data, or a conversation.

    line 72
  4. Very specific:

    Run node scripts/validate.mjs from the repo root.

    line 76
  5. Distinctive:

    Write in the user's language.

    line 84

≈ “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

Mind maker

Turns a person's documented way of working into a skill that changes what an agent does. The goal is a method, not an impression: steps, checks and output formats that someone could follow without knowing the person.

1. Consent and scope check (do this first)

  • Living person: use only principles they published or stated publicly, cited. No imitation of voice, catchphrases, or personality. Add the line "Not affiliated with or endorsed by <name>." If the person is a private individual, a colleague, or a creator with a community (for example a YouTuber or streamer), ask the user whether they have the person's permission. Without it, stop and suggest building the skill around the method itself, without the person's name.
  • Do not build: politicians, religious leaders, royalty, or anyone where the skill could be read as speaking for them on public affairs.
  • Historical figures: a light voice is fine, but the method still comes first and quotes still need sources.

2. Gather and read sources

Ask for, or locate, 2–5 primary sources: the person's own books, papers, talks, letters, posts. Secondary sources (biographies, articles) only to fill gaps, marked as such. For each source, note what it says about how they work, not what they achieved.

3. Extract

Fill this table before writing anything:

PartQuestions
Distinct jobWhat does this method do that no existing mind in the pack already does? If it overlaps heavily, stop and say which mind covers it.
MethodWhat steps do they actually follow? In what order? What do they check?
OutputWhat artifact or format does their way of working produce?
Anti-patternsWhat do they explicitly reject or warn against?
TriggersIn what situations would a user want this mind? Phrases in English and in the user's language.
QuotesAny exact quote must have a source link. Otherwise paraphrase.

4. Write the skill

Create skills/<slug>/SKILL.md with this structure (match the other minds):

---
name: <slug>
description: >
  <who> <method in one sentence>. Use when <situations>. Triggers: "<phrase>", "<phrase>", "<phrase in user's language>".
license: MIT
metadata:
  mind: <Full Name> (<years>)
  living: "<true|false>"
---

# Think like <Name>

<2–3 lines: who, and why their method matters for this kind of work.>
<If living: Not affiliated with or endorsed by <Name>. Built from <sources>.>

## When to use
## Method
## Output
## Anti-patterns
## Example (illustration)
## Sources

Keep it 60–120 lines. Method steps must be things an agent can do with code, docs, data, or a conversation. The example is labelled as an illustration and uses a realistic engineering task.

5. Validate

Run node scripts/validate.mjs from the repo root. Fix every error. Then try the skill on one real task and check that its output follows the format.

Output to the user

  • The consent/scope decision
  • The extraction table
  • The new SKILL.md
  • Validator result Write in the user's language.

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.

Check your setup, free