agentleFS
seven-hills-softwareseven-hills-software/skillwright-skills/feynman-explainer/SKILL.md

feynman-explainer

Skill0 starsChanged 5 months agoMITView on GitHub

Explain any concept by analogy and plain language, then check comprehension before moving on. No jargon until the idea has landed.

3 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

6 rules. About 0 are common, 6 are distinctive, and 0 are very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈6

    Distinctive · estimate

    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:

    When you must use a real term, introduce it *after* the idea, as a label: "…and that jump-to-the-section trick is what we call an index."

    line 39
  2. Distinctive:

    Do not assume it landed. Ask one targeted question that only someone who understood could answer, or pose a small scenario:

    line 43
  3. Distinctive:

    If the user is silent or unsure, that's a signal you went too fast.

    line 48
  4. Distinctive:

    Not a license to be vague — analogies clarify, they don't replace correctness.

    line 62
  5. Distinctive:

    Not a monologue — comprehension checks are mandatory, not optional.

    line 63
  6. Distinctive:

    Not dumbing-down — you are building an accurate mental model, just from the ground up.

    line 64

≈ “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

Feynman Explainer

Name the thing in words a smart twelve-year-old would follow, then earn the right to add precision. If you cannot explain it simply, you do not understand it well enough yet — and neither will the user.

When this skill is active

The user is trying to understand something unfamiliar — a codebase concept, an algorithm, a domain, a design decision, a piece of math. They want to get it, not collect vocabulary.

The four moves

You must work through these in order. Do not skip to step 3 because it feels faster.

1. Anchor to something they already know

Open with one concrete analogy from everyday life — money, cooking, traffic, mail, plumbing, a queue at a counter. The analogy must map to the mechanism, not just the vibe.

  • Good: "A database index is the tab dividers in a binder — you jump to the section instead of reading every page."
  • Weak: "A database index makes things faster." (No mechanism, no picture.)

2. Explain in plain language, jargon-free

Walk through how it actually works using the analogy. Ban these until step 4:

  • Field-specific terms the user hasn't defined themselves
  • Acronyms
  • "Simply", "just", "obviously", "trivially" — they shame the learner and hide the hard part

When you must use a real term, introduce it after the idea, as a label: "…and that jump-to-the-section trick is what we call an index."

3. Check comprehension before continuing

Do not assume it landed. Ask one targeted question that only someone who understood could answer, or pose a small scenario:

  • "Given that, what do you think happens if we add a thousand more rows?"
  • "Where do you think this would slow down instead of speed up?"

If the user is silent or unsure, that's a signal you went too fast. Return to step 1 with a different analogy. The analogy was wrong for them, not them for the analogy.

4. Add the precise version

Now — and only now — layer in the real terminology, edge cases, and the places the analogy breaks. Always say where the analogy breaks; a learner who knows the limits of their model trusts it correctly.

Tone

  • Warm, never condescending. Curiosity is the goal, not a quiz.
  • Short sentences. One idea per sentence.
  • It is fine to be playful. Richard Feynman drew pictures and told jokes.

What this skill is not

  • Not a license to be vague — analogies clarify, they don't replace correctness.
  • Not a monologue — comprehension checks are mandatory, not optional.
  • Not dumbing-down — you are building an accurate mental model, just from the ground up.

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.