Science Concept Explanation Skill

A science explainer packaged as a Claude Skill. Not a Wikipedia dump — layered intuition calibrated to your audience level, two analogies from different domains with their breaking points named explicitly, a diagram detailed enough to sketch, preempted misconceptions, and a comprehension check that can only be passed with real understanding.

What the skill does

• Audience-calibrated explanation. Pick your level — 12-year-old / non-science major / someone who knows basic physics / advanced student / graduate researcher — and the skill calibrates every section accordingly. No condescension, no over-reach.
• Intuition before formalism. Every concept starts with a familiar phenomenon at your level, surfaces the question it raises, then shows how the concept answers it. The formal definition comes later, not first.
• Two analogies from different domains. One mechanical + one biological, or one everyday-object + one macroscopic — never two from the same category. Each analogy is mapped element-by-element to the concept, and the skill names exactly where the analogy breaks down if pushed too far.
• Sketchable text-described diagram. Every element labeled, spatial relationships precise, axes and units where applicable — detailed enough that you could draw it from the description alone.
• Three real-world uses from different domains (not three from physics), each in two sentences — so the concept lands as a tool, not a trivia entry.
• Three preempted misconceptions. For each: state it, explain why it's intuitive, then state the correct picture with the observation or experiment that forces the correction. Famous pop-science misrepresentations (quantum anything-can-happen, we-only-use-10%-of-our-brains, evolution-as-survival-of-the-fittest) are called out directly.
• Comprehension check with tiered hints. One question that can only be answered correctly if you actually understood (not pattern-matched from the explanation), plus three tiers of hint — nudge / partial / full — revealed only on request.
• Three "going deeper" directions — one that adds rigor, one connected to an adjacent concept you probably know, one at a research frontier or open question.
• Interactive commands: SIMPLER / DEEPER (recalibrate one level), MATH (pull in the equations), HISTORY (who discovered it and how), EXPERIMENT (the key experiment that established it), COMPARE <other concept> (side-by-side table when two concepts feel interchangeable), QUIZ (5 items from what you've covered).
• Accuracy over simplicity. If a simplification would mislead, the skill says "the simplified picture is X, but the more accurate statement is Y" and explains the gap. Units attached to numbers; Greek letters named on first use.

How it works

1. Intake — on first run the skill introduces itself, asks the concept you want explained and your target audience level.
2. Explanation pass — produces all eight sections in order: elevator definition → intuition → two analogies → diagram → real-world uses → misconceptions → comprehension check → going deeper.
3. Comprehension loop — you answer the comprehension question in your own words. If you miss, tiered hints let you self-correct rather than being handed the answer.
4. Depth adjustment — SIMPLER / DEEPER recalibrates if the level feels off, COMPARE resolves confusion between related concepts, MATH / EXPERIMENT / HISTORY pulls the relevant deeper thread.

How to use it

1. Click ⬇ Download this Claude Skill above.
2. Import the .md file — either through Claude Desktop (Customize → Skills → + → Create skill → Upload a skill) or by dropping it into .claude/commands/ or ~/.claude/commands/ for Claude Code. Full walkthrough in the import tutorial.
3. Invoke the skill with the concept and your level:

   /science-concept-explanation entropy, non-science major
   

4. Answer the comprehension check in your own words before reading the hints — that's where understanding forms.

---

Quick-start prompt (no download)

Prefer a one-shot explanation without installing anything? Paste this into Claude:

Explain [scientific concept] to me as if I were [12-year-old / non-science major / someone who knows basic physics / advanced student / graduate researcher]. Produce, in order:

1. A one-sentence elevator definition — the shortest true statement calibrated to my level.
2. Intuition build-up — start with a familiar phenomenon at my level, identify the question it raises, then show how the concept answers the question. Do not start with the formal definition.
3. Two analogies from different domains (one mechanical + one biological, or one everyday-object + one macroscopic — not two from the same category). For each: state it, map it element-by-element to the concept, and name exactly where the analogy breaks down if pushed too far.
4. A text-described diagram detailed enough that I could sketch it — every element labeled, spatial relationships precise, axes and units if applicable.
5. Three real-world uses from different domains (not three from physics) — each in two sentences.
6. Three common misconceptions — for each, state it, explain why it's intuitive, and state the correct picture with the observation or experiment that forces the correction.
7. One comprehension question that can only be answered correctly if I actually understood (not pattern-matched from the explanation). Keep 3 tiers of hint ready — nudge / partial / full — reveal only on request.
8. Three "going deeper" directions — one that adds rigor, one connected to an adjacent concept I probably know, one at a research frontier or open question.

Never sacrifice accuracy for simplicity. If a simplification would mislead, say "the simplified picture is X, but the more accurate statement is Y" and explain the gap. Preempt famous pop-science misrepresentations (quantum anything-can-happen, we-only-use-10%-of-our-brains, evolution-as-survival-of-the-fittest). Attach units to numbers; name Greek letters on first use.

Tips:

• The downloadable skill is the better path if you're exploring a topic over several sessions — COMPARE and DEEPER become much more useful when the skill remembers what you've already covered.
• Answer the comprehension check in your own words before reading the hints — that is where understanding actually forms.
• Ask COMPARE whenever two concepts feel interchangeable; the table will show exactly when each applies.
• If the explanation feels either condescending or over-your-head, ask for SIMPLER or DEEPER — recalibration is one command away.

⚠ This skill has been tested and optimized for Claude. Results may vary with other AI assistants.