Sign inGet started
Skills / Learning

The Feynman Technique

I learned very early the difference between knowing the name of something and knowing something.

The Feynman Technique is a four-step learning framework that uses the act of teaching as a diagnostic tool. By forcing yourself to explain a concept in plain language — as if to a curious 12-year-old — you reveal exactly where your understanding is shallow, circular, or hiding behind jargon. Repeat the loop until your explanation is clean, gap-free, and anchored in a concrete analogy. Named after physicist Richard Feynman, who was as legendary for making hard science feel inevitable as for winning the Nobel Prize.

By Richard Feynman · Free
Specimen 01 · Live diagnosisThe Feynman Technique
Input

“The Central Limit Theorem. I have a statistics exam in three days. I've read the textbook section twice but I'm not sure I actually get it — I might…”

Diagnosis
Welcome.
Full transcript ↓
Calibrated referenceagent-skills.ai
The gap

Expose every gap in your understanding by teaching what you think you know

The Feynman Technique is a four-step, iterative framework for building genuine understanding rather than surface familiarity. Step 1 (Choose and Study): Select a concept and study it with the explicit goal of being able to teach it. Step 2 (Blank-Page Explanation): Write out a plain-language explanation as if teaching someone with no background — in your own words, no jargon. Step 3 (Gap Identification): Wherever the explanation stutters, becomes vague, or relies on unexplained technical terms, you have found a knowledge gap. Return to source material and study specifically those sub-concepts. Step 4 (Simplify and Add Analogies): Re-examine the full explanation, simplify further, and replace complex language with analogies that map the concept onto familiar experiences. Steps 2–4 repeat until the explanation is clean, gap-free, and genuinely simple. The technique treats jargon as a diagnostic red flag and analogy-construction as the measure of mastery.

The problem

Most studying produces the illusion of competence. Re-reading notes feels productive, and a highlighted textbook feels like understanding — until you are asked to explain the concept without your materials and the words dry up. The problem is not effort; it is the method. Passive re-reading reinforces familiarity with vocabulary, not command of mechanisms. The Feynman Technique is designed to break this illusion: the moment you must teach something, you discover whether you understand the concept or just its name.

The solution

After one session, you will have a clean plain-language explanation of your concept, a precise list of every gap that was hiding beneath familiar-sounding language, and at least one analogy that makes the concept feel mechanically intuitive — not just memorable. You will know exactly what to study next, and exactly how well you understand it now.

You bring
  • A concept, topic, or subject you are currently studying or want to understand deeply
  • Any notes, textbook passages, or source material you have already read (optional but useful)
  • A first-draft explanation of the concept if you have already attempted it (optional)
  • Honest acknowledgment of which parts feel fuzzy or are borrowed from memory rather than understood
You get
  • A plain-language explanation of your concept that would make sense to a non-expert — written in your own words, jargon-free
  • A precise Knowledge Gap Map identifying the exact passages where your understanding was shallow, circular, or jargon-dependent
  • Targeted study questions for each identified gap, so your re-study is active and focused rather than passive re-reading
  • A Feynman Summary Card: your concept on one page — plain explanation, gaps found and resolved, and a concrete working analogy
In action

Watch the methodology work.

Three specimens from a single real session: the same situation, unaided and calibrated, the full transcript, and the skill answering live in the channel where the work happens.

Fig.01The same situation, two ways
Specimens A · B
Specimen AVague
Unaided

“You have read the thermodynamics chapter twice. You feel like you understand entropy — you can spot the word, you've highlighted the textbook definition, and you could probably answer a multiple-choice question. Then your study partner asks you to explain it without using the words 'disorder' or 'randomness.' You stall. You realize you have been memorizing vocabulary, not understanding why heat flows in one direction and not the other.”

×Vague — nothing concrete to act on.
Specimen BCalibrated
With The Feynman Technique

“You ran the Feynman Technique on entropy. Your first blank-page explanation revealed two gaps: you didn't actually know why heat flows from hot to cold rather than the reverse, and your 'disorder' definition was circular — you were defining disorder using disorder. Forty minutes of targeted re-study and one iteration later, you have a plain-language explanation that holds up without jargon, a probability analogy for why the high-entropy state is overwhelmingly more likely, and a Summary Card ready for exam morning. More importantly, you can field any follow-up question — because you understand the mechanism, not just the name.”

Specific, grounded, and ends with one clear next step.
Fig.02A real session, unedited
In-app · assistant
The Feynman TechniqueClaude
The Feynman Technique
Welcome. The Feynman Technique runs on a single premise: you cannot fake understanding when you have to teach. We will follow all four steps — blank-page explanation, gap identification, targeted re-study, and analogy construction. What concept do you want to understand deeply today?
The Central Limit Theorem. I have a statistics exam in three days. I've read the textbook section twice but I'm not sure I actually get it — I might just be recognizing the words.
Two textbook reads is precisely the kind of studying the Feynman Technique is designed to test. Here is what we are going to do: forget everything you have highlighted. Explain the Central Limit Theorem right now as if you are teaching a curious 15-year-old who has never taken a statistics course. Use your own words. No technical terms unless you immediately explain what they mean in plain language. Go ahead.
Okay: The Central Limit Theorem says that if you take a large enough sample from any distribution, the sampling distribution of the mean will be approximately normally distributed. The larger the sample size, the closer it gets to a normal distribution, regardless of what the underlying population distribution looks like.
Good attempt — and the Jargon Gap Scanner just flagged three specific gaps in that explanation:
Gap 1 — 'sampling distribution of the mean': You used this term without explaining it. A 15-year-old would ask: what IS a sampling distribution? What are you taking the mean of, and why would you do that repeatedly?
Gap 2 — 'normally distributed': You assumed the reader knows what this shape is and why it matters. What does it look like? Why does the shape matter?
Gap 3 — 'regardless of what the underlying population distribution looks like': This is the surprising and important claim — and you stated it without any mechanism. Why does the shape of the original population stop mattering?
Gap 3 is where most explanations are shallowest. Let's go there first. In your own words — why does averaging many samples smooth out the original shape of the population?
I think it's because when you average a lot of numbers, the extreme values cancel out? The high values and low values balance each other, so the average ends up near the middle. But I'm not sure why that specifically produces a bell curve shape.
That is exactly the right mechanism — you have it. The averaging crushes extremes because high outliers and low outliers cancel. Now let's build the analogy that makes this stick rather than float. Three candidates:
Analogy A (Dice): Roll one die — any number 1 through 6 is equally likely, a flat distribution. Now roll 10 dice and average them. You almost never get 1 or 6 — most averages land near 3.5. Roll 100 dice: the average is nearly always between 3.3 and 3.7. The averaging demolishes the extremes and produces a bell curve centered on 3.5.
Analogy B (Crowd heights): Measure one random person — could be anyone. Now measure the average height of randomly drawn groups of 30 people. Those group averages cluster tightly around the population average, even though individual heights vary widely. The bigger the group, the tighter the cluster.
Analogy C (Measurement noise): Each measurement of a physical quantity has random error — sometimes high, sometimes low. Average many measurements and the high errors and low errors cancel, leaving you close to the true value. The distribution of those averages forms a bell around the truth.
Which one gives you the clearest mental image?
The dice one — I can almost visualize it. Roll a hundred dice and the average is always near 3.5, even though any single die is totally unpredictable.
That visualization is your understanding working correctly. Here is your Feynman Summary Card:
---FEYNMAN SUMMARY CARD: Central Limit Theorem
Plain Explanation: When you draw many samples from any population and calculate each sample's average, those averages form a bell curve centered on the true population average — regardless of the shape of the original population. The larger each sample, the tighter the bell curve becomes.
Gaps Found and Resolved:- Sampling distribution: the pattern you get by repeating 'draw a sample, calculate its mean' many times- Normal distribution: the bell-curve shape — most values near the center, thinning symmetrically toward the edges- Why shape stops mattering: extreme values in a sample cancel each other during averaging, so averages cluster in the middle regardless of the original distribution's shape
Core Analogy: Roll 100 dice and average the result — you will almost always land near 3.5, no matter how random any single roll is. Scale up the sample size, and the average becomes nearly certain.
---
You went from recognizing words to understanding a mechanism. That is what the technique is for.
Reply to The Feynman Technique
What the skill is doing
01Concept Teach-Back
02Jargon Gap Scanner
03Targeted Restudy Prompts
Fig.03The same skill, in the channel you already use
WhatsApp · live
9:41
TFThe Feynman Techniquetyping…
Messages are end-to-end encrypted. No one outside of this chat can read them.
Today
The Central Limit Theorem. I have a statistics exam in three days. I've read the textbook section twice but I'm not sure I actually get it — I might j…9:14✓✓
The Feynman Technique runs on a single premise: you cannot fake understanding when you have to teach.9:14
We will follow all four steps — blank-page explanation, gap identification, targeted re-study, and analogy construction.9:15
What concept do you want to understand deeply today?9:15
Message

The same skill, where the work happens.

No new app to learn. The methodology runs over the WhatsApp Business API, so the answer lands as a reply in the thread you’re already in — same rigour, zero context-switch.

Reads the situation, names the pattern, returns one concrete next move.
Delivered in seconds, inside a conversation that already exists.
Specimen · WhatsApp Business API · live
Capabilities

What it does, specifically.

Each capability is a distinct move drawn straight from the source methodology — not a generic assistant guessing.

CapabilityC-01

Concept Teach-Back

You introduce a concept and the skill guides you through drafting a plain-language explanation from a blank page — no copying from notes, no technical shorthand. The act of retrieval and reformulation reveals what you actually understand versus what you have only read. The skill holds the frame: plain language, your own words, no jargon allowed.

Based on Feynman's Step 2: write the explanation as if teaching someone with no background in the subject, using entirely your own words — the core diagnostic step that separates familiarity from understanding.
CapabilityC-02

Jargon Gap Scanner

The skill reads your explanation and flags every passage that relies on unexplained technical terms, circular definitions (where X is defined using X), or vague language that signals shallow understanding. Each flagged passage is named as a specific knowledge gap, not a general 'study more' verdict.

Based on Feynman's Step 3 principle: jargon is a diagnostic red flag for gaps — the technique mandates identifying precisely where explanation breaks down before returning to source material.
CapabilityC-03

Targeted Restudy Prompts

For each knowledge gap identified, the skill generates a specific study question — not 'review chapter 4' but 'what actually happens at the molecular level when entropy increases? Explain the mechanism, not just the direction.' This makes re-study active rather than passive, and focused rather than broad.

Based on Feynman's iterative loop after gap identification: return to source material with a precise question to answer, not an unfocused second read of the same pages.
CapabilityC-04

Analogy Workshop

Once the core explanation is gap-free, the skill proposes 2–3 concrete analogies that map the concept onto familiar experiences. You refine, reject, or extend them until one feels genuinely clarifying — not just clever. A working analogy that you can extend and probe is the hallmark of Feynman-style mastery.

Based on Feynman's Step 4: add analogies that map unfamiliar concepts onto familiar experiences — Feynman's own teaching was legendary for this, such as explaining electrical resistance using water pressure through pipes of different widths.
CapabilityC-05

Feynman Summary Card

At the end of a session, the skill produces a structured one-page summary: your plain-language explanation, the specific gaps that were found and how they were resolved, and the core analogy. The Summary Card is a transformed understanding artifact — not your original notes, but evidence that you can teach the concept.

Based on the tangible output the Feynman Technique mandates: a clean, gap-free explanation with analogies that demonstrates genuine understanding — the paper Feynman reportedly kept for each concept he was studying.
Tested

Graded before it shipped.

Every skill is scored against independent scenarios for methodology fidelity before it goes live — not vibes, a rubric.

What it produces
OutputD-01

Feynman Summary Card

A one-page structured output containing your refined plain-language explanation, the specific gaps that were identified and resolved during the session, and the core working analogy — everything needed to demonstrate or recall genuine understanding without your source materials.

OutputD-02

Knowledge Gap Map

A precise list of the exact points where your initial explanation relied on jargon, circular reasoning, or vague language, with each gap named and linked to a targeted study question for focused re-study rather than unfocused re-reading.

OutputD-03

Analogy Portfolio

Two to three candidate analogies generated for your concept, each mapping it onto a different familiar domain. The strongest analogy becomes part of your Summary Card; the others offer alternative entry points for different audiences or contexts.

The source

Grounded in the original work.

Every answer traces back to a real source and the practitioner who wrote it — not a secondhand summary. Here is the source of record.

Source authorA-01

Richard Feynman

Richard Feynman (1918–1988) was an American theoretical physicist who won the Nobel Prize in Physics in 1965 for his work on quantum electrodynamics. He held a professorship at Caltech for three decades, developed Feynman diagrams, and contributed to the Manhattan Project. Feynman was as celebrated for his extraordinary gift for teaching as for his research — his introductory physics lectures drew overflow crowds, and his autobiography 'Surely You're Joking, Mr. Feynman!' became an international bestseller. He was widely known as 'The Great Explainer.'

Status · Inspired by Richard Feynman’s work — not yet claimed. Are you Richard Feynman?
Primary sourceS-01

Surely You're Joking, Mr. Feynman! / The Feynman Lectures on Physics

by Richard Feynman

Nobel Prize in Physics (1965); Caltech professor; developer of Feynman diagrams; author of 'The Feynman Lectures on Physics'; widely regarded as the greatest science communicator of the 20th century.

Read the original ↗
Citationen.wikipedia.org
Get started

Put it to work.

agent-skills runs as MCP — connect it once and The Feynman Technique works in Claude, ChatGPT, Cursor, or any MCP-compatible client. Sign in to see your config and accept the terms.