Simplify topic
Skill yugash007/edu-agent-skills/skills/teaching/simplify-topic
Reusable educational skills for AI coding agents. Turn agents (Gemini CLI, Claude Code, Cursor, etc.) into Socratic mentors and active learning companions with a single command.
npx -y skills add yugash007/edu-agent-skills --skill simplify-topicAssembled from the repository path, not quoted from the project. Check it against their README if it does not work.
One thing to look at
- 4 stars4 stars. Stars are a popularity signal and not a quality one, but at this level it is likely that nobody has read this closely except its author, and you would be relying on your own review.
What its author says it does
Copied from the file, not written here
Use when a concept is too abstract for the learner's current level and needs analogy-first teaching before complexity re-injection.
SKILL.md
3.9 KB, as published. Nobody here has run it
Purpose
Reframe a concept at a lower abstraction so the learner builds accurate intuition before encountering technical complexity. Simplification is choosing the right abstraction for the learner's mental model, then re-injecting complexity once the foundation is stable.
Activation
- Learner says "I don't understand" or "explain like I'm new."
check-understandingreveals prior explanation was too abstract/jargon-heavy. Cross-domain learner needs an analogy bridge.teach-conceptwas used and learner still can't state the core intuition. - Skip if: learner understands basics and needs depth →
deep-dive. Confusion is from a misconception →misconception-detector. Concept is simple enough already. - Routing: after successful simplification, use
teach-conceptto re-introduce correct technical terminology. Never leave learner permanently at simplified level.
Inputs
- Concept to simplify, learner's background domain, abstraction level where understanding broke, specific unclear aspect.
Abstraction Levels
- Level 0 (Everyday): household/daily life analogy, zero technical vocab → absolute beginner.
- Level 1 (Domain-Adjacent): analogy from learner's known field → practitioner switching domains.
- Level 2 (Simplified Technical): correct terms, simplified mechanism → beginner with some background.
- Level 3 (Full Technical): precise mechanism with edge cases → standard
teach-conceptlevel.
Workflow
- Detect — Ask one question to identify where understanding breaks (terminology? mechanism? motivation?). Identify learner's domain for analogy selection. Select starting level (0, 1, or 2).
- Build Analogy — One analogy from the learner's known domain. State explicitly where it holds AND where it breaks down — oversimplified analogies create new misconceptions.
- Simplified Explanation — Deliver at selected level using the analogy as scaffold. One core idea per step. No technical vocabulary at level 0; introduce terms one at a time at level 2.
- Intuition Check — Ask learner to restate in their own words using the analogy. If wrong: identify failure point and rebuild with a different analogy.
- Complexity Re-Injection — Introduce one layer of technical accuracy on top of confirmed intuition. Replace analogy language with correct terms, one at a time. Confirm at each step.
- Handoff — When learner articulates using correct vocabulary: hand off to
teach-conceptordeep-dive.
Rules
- DO: always state where the analogy breaks down — analogies that never fail create misconceptions.
- DO: complexity re-injection is mandatory — simplification is a bridge, not a destination.
- DO: confirm the analogy domain is familiar to the learner before using it.
- DON'T: use multiple analogies simultaneously — pick the best one until it fails.
- DON'T: discard a failing analogy until after 2 attempts; then switch.
- DON'T: leave the session at simplified level — at minimum introduce correct vocabulary.
- DON'T: simplify a concept the learner already understands correctly — probe first.
Output
Responses should contain: context (concept + confusion level + learner background + selected level), analogy (with explicit limits), simplified explanation, intuition check prompt, complexity re-injection steps, and handoff plan. Format naturally.
Checklist
- Analogy grounded in learner's known domain and limits stated.
- Learner restates concept using analogy before re-injection.
- Complexity re-injection introduces correct technical vocabulary.
- Session does not end at simplified level.