teach-me
Use when the user wants to deeply learn a new topic from scratch. Runs a pre-interview (current knowledge, end-goal proficiency, depth, practice load, background, scope), researches online (articles, niche-influencer blogs, canonical docs, subtopic landscape), then produces a structured markdown course with mandatory visual diagrams, evidence-based learning-science features (retrieval practice, spaced callbacks, worked-example fading, concept ledger, jargon gate, analogy hygiene), and a self-contained interactive HTML mini-course. Triggers on /teach-me, "teach me about X", "I want to learn X", "deep dive on X", "create a course on X", "study X with me".
AlemTuzlak · 已有 AI 审核记录
安装与使用
运行命令前先阅读来源与依赖说明。复制命令不会自动安装或执行。
npx skills add AlemTuzlak/skills --skill teach-me技能说明
name: teach-me description: Use when the user wants to deeply learn a new topic from scratch. Runs a pre-interview (current knowledge, end-goal proficiency, depth, practice load, background, scope), researches online (articles, niche-influencer blogs, canonical docs, subtopic landscape), then produces a structured markdown course with mandatory visual diagrams, evidence-based learning-science features (retrieval practice, spaced callbacks, worked-example fading, concept ledger, jargon gate, analogy hygiene), and a self-contained interactive HTML mini-course. Triggers on /teach-me, "teach me about X", "I want to learn X", "deep dive on X", "create a course on X", "study X with me".
/teach-me — Evidence-Based Course Generator
Produce a self-paced markdown course + self-contained HTML mini-course for any topic the user wants to learn. Designed for adult learners (CEFR B1 reading level, no prior knowledge), grounded in cognitive psychology and instructional-design research.
When to Use
- User invokes
/teach-me <topic>(or no arg — ask for the topic) - User says "teach me about X", "I want to learn X", "deep dive on X", "create a course on X", "study X with me"
- Skip for: quick factual questions ("what is X" — just answer), code help, debugging, doc lookups, library API questions (use
context7instead)
Pedagogy Foundation (read before starting work)
This skill is grounded in well-replicated learning science. The rules below are not negotiable defaults — they are what the user is paying for:
- Concrete before abstract (Goldstone, Fyfe et al.). Every chapter opens with a concrete observable instance before any abstract definition.
- Retrieval practice beats re-reading (Roediger & Karpicke; Rowland 2014 meta-analysis). Every chapter ends with 3–5 free-recall prompts + 1 cross-chapter callback.
- Spaced exposure via deliberate callbacks (Cepeda et al. 2006). Each major concept is used (not redefined) in ≥2 later chapters.
- Worked examples + backward fading (Sweller, Renkl & Atkinson). Early chapters: full worked examples. Middle: progressively-blanked steps. Late: prompt-only.
- Cognitive Load Theory (Sweller, Mayer). Max ~4 new named concepts per chapter. One new dimension of difficulty at a time.
- Segmenting / cards (Mayer & Pilegard). Chapters are decks of 250–500-word cards, one concept per card, one diagram per card.
- Diagram-first visual coherence (Mayer multimedia principles). Mandatory diagrams for relational concepts. No decorative imagery. Labels inline on the diagram, not in separate legends.
- Analogy hygiene (Gentner structure-mapping). Every analogy followed within 2 sentences by an explicit "where this breaks down" disclaimer.
- Jargon gate (Nathan/Koedinger expert-blind-spot research). Every technical term defined on first use OR linked to a glossary entry from an earlier chapter. No exceptions for "common" terms.
- Constrained self-explanation prompts (Chi et al.; Bisra et al. 2018 g≈0.55). Between cards: "Finish this sentence: X works because ___" with model answer in a toggle.
- Plain language for B1 readers. Mean sentence ≤18 words, max 25. ≤1 subordinate clause per sentence. Passive ≤10%. Active voice. Second person. No idioms.
- Build on what the learner already knows (Ausubel). The pre-interview captures the user's known domains; the writing guide turns them into an analogy bank.
- Show state changes; don't narrate them (Larkin & Simon 1987; Tversky & Morrison 2002 Apprehension Principle; Wong/Leahy/Marcus/Sweller 2012 transient-information effect; Tufte small multiples; Mayer's segmenting). When prose would walk through successive states ("first this, then that, after the crash…"), use a 2–5 panel
<div class="state-sequence">instead. Same shape in every panel, delta annotated, caption = trigger + invariant. The reader sees the evolution; they don't read a story about it.
Debunked — never use: learning styles (VAK), 10,000-hours framing, gamification streaks/badges, decorative imagery for "engagement", "right-brain/left-brain", "digital natives".
Process Flow
digraph teach_me {
rankdir=TB;
"Parse invocation" [shape=box];
"Existing folder?" [shape=diamond];
"Output location" [shape=box];
"Pre-interview" [shape=box];
"Research (4 parallel agents)" [shape=box];
"Synthesize + write OUTLINE.md" [shape=box];
"User picks chapters" [shape=box];
"Write WRITING_GUIDE.md" [shape=box];
"Chapter generation (parallel batches)" [shape=box];
"Consistency pass" [shape=box];
"Build index.html" [shape=box];
"Open folder + browser" [shape=doublecircle];
"Parse invocation" -> "Existing folder?";
"Existing folder?" -> "Output location" [label="no"];
"Existing folder?" -> "Output location" [label="extend / fresh / cancel"];
"Output location" -> "Pre-interview";
"Pre-interview" -> "Research (4 parallel agents)";
"Research (4 parallel agents)" -> "Synthesize + write OUTLINE.md";
"Synthesize + write OUTLINE.md" -> "User picks chapters";
"User picks chapters" -> "Write WRITING_GUIDE.md";
"Write WRITING_GUIDE.md" -> "Chapter generation (parallel batches)";
"Chapter generation (parallel batches)" -> "Consistency pass";
"Consistency pass" -> "Build index.html";
"Build index.html" -> "Open folder + browser";
}
Do not skip phases. Track progress with TaskCreate so the user sees where you are.
Phase 0 — Parse invocation
- Get the topic from arguments. If no arg, reply: "What topic do you want to learn? It can be technical (a library, language, concept) or non-technical (a philosophy, field, skill)." and wait.
- Topic validation guards:
- If the topic is harmful (weapon synthesis, malicious hacking targets, illegal acts) — refuse briefly and stop. No lecture.
- If the topic is impossibly broad (
programming,science,everything) — ask the user to narrow with 2–3 concrete examples: *"That's very broad