Initial commit

skills/socratic-teaching-scaffolds/SKILL.md

@@ -0,0 +1,245 @@
+---
+name: socratic-teaching-scaffolds
+description: Use when teaching complex concepts (technical, scientific, philosophical), helping learners discover insights through guided questioning rather than direct explanation, correcting misconceptions by revealing contradictions, onboarding new team members through scaffolded learning, mentoring through problem-solving question frameworks, designing self-paced learning materials, or when user mentions "teach me", "help me understand", "explain like I'm", "learning path", "guided discovery", or "Socratic method".
+---
+
+# Socratic Teaching Scaffolds
+
+## Table of Contents
+1. [Purpose](#purpose)
+2. [When to Use](#when-to-use)
+3. [What Is It](#what-is-it)
+4. [Workflow](#workflow)
+5. [Socratic Question Types](#socratic-question-types)
+6. [Scaffolding Levels](#scaffolding-levels)
+7. [Common Patterns](#common-patterns)
+8. [Guardrails](#guardrails)
+9. [Quick Reference](#quick-reference)
+
+## Purpose
+
+Socratic Teaching Scaffolds guide learners to discover knowledge through strategic questioning and progressive support removal. This skill transforms passive explanation into active discovery, corrects misconceptions by revealing contradictions, and builds durable understanding through self-generated insights.
+
+## When to Use
+
+**Invoke this skill when you need to:**
+- Teach complex concepts where understanding beats memorization (algorithms, scientific theories, philosophical ideas)
+- Correct deep misconceptions that resist direct explanation (statistical fallacies, physics intuitions, programming mental models)
+- Help learners develop problem-solving skills, not just solutions
+- Design learning experiences that build from concrete to abstract understanding
+- Onboard professionals through guided discovery rather than documentation dumps
+- Create self-paced learning materials with built-in feedback loops
+- Mentor through questions that develop independent thinking
+- Bridge expertise gaps across different knowledge levels
+
+**User phrases that trigger this skill:**
+- "Teach me [concept]"
+- "Help me understand [topic]"
+- "Explain like I'm [expertise level]"
+- "I don't get why [misconception]"
+- "What's the best way to learn [skill]?"
+- "How should I think about [concept]?"
+
+## What Is It
+
+A teaching framework combining **Socratic questioning** (strategic questions that guide discovery) with **instructional scaffolding** (temporary support that fades as competence grows).
+
+**Core components:**
+1. **Question Ladders**: Sequences from simple to complex that build understanding incrementally
+2. **Misconception Detectors**: Questions that reveal faulty mental models through contradiction
+3. **Feynman Explanations**: Build-up from simple analogies to technical precision
+4. **Worked Examples with Fading**: Full solutions → partial solutions → independent practice
+5. **Cognitive Apprenticeship**: Model thinking process explicitly, then transfer to learner
+
+**Quick example (Teaching Recursion):**
+
+**Question Ladder:**
+1. "Can you break this problem into a smaller version of itself?" (problem decomposition)
+2. "What would happen if we had only one item?" (base case discovery)
+3. "If we could solve the small version, how would we use it for the big version?" (recursive case)
+4. "What prevents this from running forever?" (termination reasoning)
+
+**Misconception Detector:**
+- "Will this recursion ever stop? Trace it with 3 items." (reveals infinite recursion misunderstanding)
+
+**Feynman Progression:**
+- Level 1: "Like Russian nesting dolls—each contains a smaller version"
+- Level 2: "Function calls itself with simpler input until base case"
+- Level 3: "Recursive definition: f(n) = g(f(n-1), n) with f(0) = base"
+
+## Workflow
+
+Copy this checklist and track your progress:
+
+```
+Socratic Teaching Progress:
+- [ ] Step 1: Diagnose learner's current understanding
+- [ ] Step 2: Design question ladder and scaffolding plan
+- [ ] Step 3: Guide discovery through questioning
+- [ ] Step 4: Fade scaffolding as competence grows
+- [ ] Step 5: Validate understanding and transfer
+```
+
+**Step 1: Diagnose learner's current understanding**
+
+Ask probing questions to identify current knowledge level, misconceptions, and learning goals. See [Socratic Question Types](#socratic-question-types) for diagnostic question categories.
+
+**Step 2: Design question ladder and scaffolding plan**
+
+Build progression from learner's current state to target understanding. For straightforward teaching → Use [resources/template.md](resources/template.md). For complex topics with multiple misconceptions → Study [resources/methodology.md](resources/methodology.md).
+
+**Step 3: Guide discovery through questioning**
+
+Ask questions in sequence, provide scaffolding (hints, worked examples, analogies) as needed. See [Scaffolding Levels](#scaffolding-levels) for support gradations. Adjust based on learner responses.
+
+**Step 4: Fade scaffolding as competence grows**
+
+Progressively remove hints, provide less complete examples, ask more open-ended questions. Monitor for struggle (optimal challenge) vs frustration (too hard). See [resources/methodology.md](resources/methodology.md) for fading strategies.
+
+**Step 5: Validate understanding and transfer**
+
+Test with novel problems, ask for explanations in learner's words, check for misconception elimination. Self-check using [resources/evaluators/rubric_socratic_teaching_scaffolds.json](resources/evaluators/rubric_socratic_teaching_scaffolds.json). Minimum standard: Average score ≥ 3.5.
+
+## Socratic Question Types
+
+**1. Clarifying Questions** (Understand current thinking)
+- "What do you mean by [term]?"
+- "Can you give me an example?"
+- "How does this relate to [known concept]?"
+
+**2. Probing Assumptions** (Surface hidden beliefs)
+- "What are we assuming here?"
+- "Why would that be true?"
+- "Is that always the case?"
+
+**3. Probing Reasons/Evidence** (Justify claims)
+- "Why do you think that?"
+- "What evidence supports that?"
+- "How would we test that?"
+
+**4. Exploring Implications** (Think through consequences)
+- "What would happen if [change]?"
+- "What follows from that?"
+- "What are the edge cases?"
+
+**5. Questioning the Question** (Meta-cognition)
+- "Why is this question important?"
+- "What are we really trying to understand?"
+- "How would we know if we understood?"
+
+**6. Revealing Contradictions** (Bust misconceptions)
+- "Earlier you said [X], but now [Y]. How do these fit?"
+- "If that's true, why does [counterexample] happen?"
+- "What would this predict for [test case]?"
+
+## Scaffolding Levels
+
+Provide support that matches current need, then fade:
+
+**Level 5: Full Modeling** (I do, you watch)
+- Complete worked example with thinking aloud
+- Explicit strategy articulation
+- All steps shown with rationale
+
+**Level 4: Guided Practice** (I do, you help)
+- Partial worked example
+- Ask learner to complete steps
+- Provide hints before errors
+
+**Level 3: Coached Practice** (You do, I help)
+- Learner attempts independently
+- Intervene with questions when stuck
+- Guide without giving answers
+
+**Level 2: Independent with Feedback** (You do, I watch)
+- Learner solves alone
+- Review and discuss afterwards
+- Identify gaps for next iteration
+
+**Level 1: Transfer** (You teach someone else)
+- Learner explains to others
+- Learner creates examples
+- Learner identifies misconceptions in others
+
+**Fading strategy:** Start at level matching current competence (not Level 5 by default). Move down one level when learner demonstrates success. Move up one level if learner struggles repeatedly.
+
+## Common Patterns
+
+**Pattern 1: Concept Introduction (Concrete → Abstract)**
+- Start: Real-world analogy or example
+- Middle: Formalize with terminology
+- End: Abstract definition with edge cases
+- Example: Teaching pointers (address on envelope → memory location → pointer arithmetic)
+
+**Pattern 2: Misconception Correction (Prediction → Surprise → Explanation)**
+- Ask learner to predict outcome
+- Show actual result (contradicts misconception)
+- Guide discovery of correct mental model
+- Example: "Will this float? [test with 0.1 + 0.2 in programming] Why not exactly 0.3?"
+
+**Pattern 3: Problem-Solving Strategy (Model → Practice → Reflect)**
+- Model strategy on simple problem (think aloud)
+- Learner applies to similar problem (with coaching)
+- Reflect on when strategy applies/fails
+- Example: Teaching debugging (print statements → breakpoints → hypothesis testing)
+
+**Pattern 4: Depth Ladder (ELI5 → Undergraduate → Expert)**
+- Build multiple explanations at different depths
+- Let learner choose starting point
+- Provide "go deeper" option at each level
+- Example: Teaching neural networks (pattern matching → weighted sums → backpropagation → optimization theory)
+
+**Pattern 5: Discovery Learning (Puzzle → Hints → Insight)**
+- Present puzzling phenomenon or problem
+- Provide graduated hints if stuck
+- Guide to "aha" moment of discovery
+- Example: Teaching recursion (Towers of Hanoi → break into subproblems → recursive solution)
+
+## Guardrails
+
+**Zone of proximal development:**
+- Too easy = boredom, too hard = frustration
+- Optimal: Can't do alone, but can with guidance
+- Adjust scaffolding level based on struggle signals
+
+**Don't fish for specific answers:**
+- Socratic questioning isn't a guessing game
+- If learner's reasoning is sound but reaches different conclusion, explore their path
+- Multiple valid approaches often exist
+
+**Avoid pseudo-teaching:**
+- Don't just ask questions without purpose
+- Each question should advance understanding or reveal misconception
+- If question doesn't help, provide direct explanation
+
+**Misconception resistance:**
+- Deep misconceptions resist single corrections
+- Need multiple exposures to contradictions
+- May require building correct model from scratch before dismantling wrong one
+
+**Expertise blind spots:**
+- Experts forget what was hard as beginners
+- Make implicit knowledge explicit
+- Slow down automated processes to show thinking
+
+**Individual differences:**
+- Some learners prefer exploration, others prefer structure
+- Adjust scaffolding style to learner preferences
+- Monitor for frustration vs productive struggle
+
+## Quick Reference
+
+**Resources:**
+- **Quick teaching session**: [resources/template.md](resources/template.md)
+- **Complex topics/misconceptions**: [resources/methodology.md](resources/methodology.md)
+- **Quality rubric**: [resources/evaluators/rubric_socratic_teaching_scaffolds.json](resources/evaluators/rubric_socratic_teaching_scaffolds.json)
+
+**5-Step Process**: Diagnose → Design Ladder → Guide Discovery → Fade Scaffolding → Validate Transfer
+
+**Question Types**: Clarifying, Probing Assumptions, Probing Evidence, Exploring Implications, Meta-cognition, Revealing Contradictions
+
+**Scaffolding Levels**: Full Modeling → Guided Practice → Coached Practice → Independent Feedback → Transfer (fade progressively)
+
+**Patterns**: Concrete→Abstract, Prediction→Surprise→Explanation, Model→Practice→Reflect, ELI5→Expert, Puzzle→Hints→Insight
+
+**Guardrails**: Zone of proximal development, purposeful questions, avoid pseudo-teaching, resist misconceptions, make implicit explicit

skills/socratic-teaching-scaffolds/resources/evaluators/rubric_socratic_teaching_scaffolds.json

@@ -0,0 +1,202 @@
+{
+  "name": "Socratic Teaching Scaffolds Evaluator",
+  "description": "Evaluates teaching sessions for diagnostic quality, question ladder design, scaffolding appropriateness, misconception correction, and transfer validation",
+  "criteria": [
+    {
+      "name": "Diagnostic Depth & Accuracy",
+      "weight": 1.4,
+      "scale": {
+        "1": "No diagnostic questions, assumes learner knowledge, starts teaching without assessment",
+        "2": "Basic questions ('Do you know X?') but no misconception probing, surface-level assessment",
+        "3": "Asks 3-5 diagnostic questions, identifies some gaps but misses misconceptions, reasonable baseline assessment",
+        "4": "Thorough diagnostic (5+ questions), identifies knowledge gaps and 1-2 misconceptions, assesses starting scaffolding level",
+        "5": "Exemplary: Deep diagnostic with mental model elicitation (concept mapping, predict-observe-explain, analogical reasoning probes), identifies specific misconceptions with evidence (not assumptions), maps prerequisite knowledge dependencies, determines precise ZPD starting point, documents learner's current mental model explicitly (even if flawed), tests at multiple Bloom levels (recall, application, analysis)"
+      }
+    },
+    {
+      "name": "Question Ladder Quality",
+      "weight": 1.5,
+      "scale": {
+        "1": "No progression, random questions, or just direct explanation (not Socratic at all)",
+        "2": "Questions exist but don't build on each other, no clear path from current to target understanding",
+        "3": "Basic progression (5-7 questions) from simple to complex, some logical flow but jumps in difficulty",
+        "4": "Clear progression (8-10 questions) from concrete to abstract, logical building, addresses identified gaps",
+        "5": "Exemplary: Comprehensive ladder (10+ questions) with explicit structure: concrete foundation (Steps 1-2 with analogy/real-world), pattern recognition (Steps 3-4), formalization (Steps 5-6 with terminology), edge cases (Steps 7-8), transfer (Steps 9-10 to novel context). Each question has clear purpose (not Socratic theater), builds on previous responses, branches based on learner answers, targets identified misconceptions specifically, includes both convergent (one right answer) and divergent (explore reasoning) questions"
+      }
+    },
+    {
+      "name": "Scaffolding Appropriateness & Fading",
+      "weight": 1.5,
+      "scale": {
+        "1": "No scaffolding (sink or swim) or constant full scaffolding (never fades), one-size-fits-all approach",
+        "2": "Some scaffolding but wrong level (too much for competent learner or too little when stuck), no adjustment based on responses",
+        "3": "Appropriate starting level (Level 3-4), some fading as learner succeeds but not systematic",
+        "4": "Starts at correct scaffolding level for learner's ZPD, fades progressively (Level 5→4→3), adjusts based on struggle signals",
+        "5": "Exemplary: Precise scaffolding calibration (starts at level matching diagnostic assessment, not defaulting to Level 5), explicit fading protocol (success → down one level, struggle >2min → up one level), tracks scaffolding per sub-concept (not global), uses cognitive apprenticeship progression (modeling → coaching → scaffolding → articulation → reflection → exploration), provides multiple scaffolding types (hints, partial examples, analogies, worked examples), monitors for optimal challenge (productive struggle vs frustration), makes scaffolding temporary and explicit ('I'm helping now, but you'll do this alone soon')"
+      }
+    },
+    {
+      "name": "Misconception Correction Rigor",
+      "weight": 1.4,
+      "scale": {
+        "1": "Misconceptions ignored or corrected by assertion ('No, it's actually X'), no exploration of why misconception exists",
+        "2": "Misconceptions mentioned but not deeply addressed, single counterexample without checking if correction stuck",
+        "3": "Addresses 1-2 misconceptions with prediction-observation-explanation, some follow-up but no persistence check",
+        "4": "Corrects misconceptions through contradiction (predict based on misconception → show actual outcome → guide to correct model), verifies correction with new examples",
+        "5": "Exemplary: Systematic misconception protocol for each identified error: (1) acknowledge without judgment, (2) elicit explicit prediction based on misconception, (3) show contradiction with actual outcome, (4) guide discovery of correct model through questions (not assertion), (5) reinforce with 2-3 new examples applying corrected understanding, (6) spaced retrieval later in session to check persistence (not just surface compliance). For deep misconceptions: builds correct model from scratch before dismantling wrong one, provides multiple diverse counterexamples (not dismissible as 'special cases'), uses role reversal ('How would you correct someone who thinks [misconception]?'), contextualizes historically if relevant"
+      }
+    },
+    {
+      "name": "Transfer & Deep Understanding Validation",
+      "weight": 1.3,
+      "scale": {
+        "1": "No transfer testing, only asks learner to repeat back what was taught, accepts memorization as understanding",
+        "2": "Tests with problem identical to examples (procedural recall only), no novel application",
+        "3": "Tests with modified problem (near transfer), asks for explanation in learner's words, some validation",
+        "4": "Tests near and far transfer (novel domain with analogous structure), asks for multi-level explanation (ELI5 and technical), validates understanding",
+        "5": "Exemplary: Comprehensive transfer validation across multiple dimensions: (1) Near transfer (same domain, similar problem), (2) Modified transfer (adapted problem), (3) Far transfer (different domain, analogous structure), (4) Creative transfer (novel synthesis). Feynman test: can explain at multiple levels (child, peer, expert, teacher handling misconceptions). Bloom's validation: tests remember, understand, apply, analyze, evaluate, create. Teaching test: 'How would you teach this to someone else? What would they find confusing?' Checks that all identified misconceptions are eliminated (not just ignored). Provides next learning steps with clear success criteria"
+      }
+    },
+    {
+      "name": "Zone of Proximal Development (ZPD) Calibration",
+      "weight": 1.3,
+      "scale": {
+        "1": "Consistently too easy (learner bored) or too hard (learner frustrated), no adjustment to learner signals",
+        "2": "Occasionally in ZPD but often above or below, slow to recognize struggle vs boredom",
+        "3": "Generally in ZPD, recognizes struggle and adjusts but may take 2-3 cycles, some frustration occurs",
+        "4": "Maintains ZPD well, quickly adjusts when learner stuck or bored (within 1-2 minutes), optimal challenge most of session",
+        "5": "Exemplary: Precise ZPD maintenance with explicit signal monitoring: Below ZPD (boredom, quick correct answers without thought → skip ahead, increase complexity), Optimal ZPD (engaged struggle, eventual success with hints → maintain current level), Above ZPD (frustration, wild guesses, giving up → increase scaffolding, break into smaller steps). Adjusts within 30 seconds of detecting signal change. Starts conservatively (more scaffolding) then fades aggressively. Dynamic per-concept calibration (concept A may be in ZPD while concept B needs more scaffolding). Makes ZPD explicit: 'This should be challenging but doable. Tell me if it's too easy or too hard.'"
+      }
+    },
+    {
+      "name": "Question Purpose & Authenticity",
+      "weight": 1.2,
+      "scale": {
+        "1": "Questions are guessing game ('What am I thinking?'), fishing for specific teacher-desired answer, no genuine inquiry",
+        "2": "Some purposeful questions but many are filler, learner often guessing rather than reasoning",
+        "3": "Most questions have clear purpose, but occasionally slips into guess-the-answer mode, generally authentic inquiry",
+        "4": "All questions purposeful (advance understanding or reveal misconception), honors learner's reasoning even when different from expected path",
+        "5": "Exemplary: Every question has explicit purpose (diagnostic, pattern recognition, formalization, edge case exploration, transfer testing), never guessing game (if learner's reasoning is sound but reaches different conclusion, explores their path), provides thinking time (30 seconds minimum, doesn't rush to hint), explores multiple valid approaches when they exist, acknowledges when learner's solution is better/different, asks meta-cognitive questions ('Why is this question important?' 'How would we know if we understood?'), balances convergent (one right answer) and divergent (explore reasoning) questions, models uncertainty ('I'm not sure. Let's think it through together.')"
+      }
+    },
+    {
+      "name": "Actionability & Learning Continuity",
+      "weight": 1.1,
+      "scale": {
+        "1": "Session ends abruptly, no summary or next steps, unclear what learner should do to solidify learning",
+        "2": "Vague next steps ('keep practicing'), no concrete recommendations, no assessment of what to review",
+        "3": "Provides summary and general next steps (practice problems), some guidance on what to focus on",
+        "4": "Clear summary of key insights, specific practice recommendations, identifies concepts needing more work",
+        "5": "Exemplary: Comprehensive learning continuity plan: (1) Session summary highlighting key insights discovered (not just facts learned), (2) Identified remaining gaps or concepts needing reinforcement, (3) Specific practice recommendations (with problem types, not just 'practice more'), (4) Spaced retrieval schedule (review in 1 day, 3 days, 7 days), (5) Success criteria for next session ('You'll know you've mastered this when you can [specific task]'), (6) Resources for self-directed learning (if applicable), (7) Session notes documenting what worked well, what to adjust, follow-up needed. For asynchronous materials: includes self-check with rich feedback, progressive hint revelation, question bank with varied types (recall, application, analysis, misconception checks)"
+      }
+    }
+  ],
+  "guidance": {
+    "by_learner_level": {
+      "novice": {
+        "focus": "Concrete examples, heavy scaffolding (Level 4-5), basic vocabulary, frequent validation, avoid overwhelming",
+        "diagnostic": "Assume minimal prior knowledge, focus on foundational misconceptions (correlation vs causation, common confusions)",
+        "scaffolding": "Start Level 5 (full modeling), fade slowly, many worked examples, explicit think-aloud",
+        "validation": "Near transfer only, ask for explanation in simplest terms (ELI5 level)",
+        "typical_score": "Aim for 3.5-4.0 (solid teaching with appropriate novice support)"
+      },
+      "intermediate": {
+        "focus": "Pattern recognition, moderate scaffolding (Level 3-4), introduce formal terminology, edge cases",
+        "diagnostic": "Test for domain-specific misconceptions, assess prerequisite knowledge boundaries",
+        "scaffolding": "Start Level 3-4 (coached/guided), fade to independent practice, mix worked and partial examples",
+        "validation": "Near and far transfer, multi-level explanations (peer and expert levels), analysis questions",
+        "typical_score": "Aim for 4.0-4.5 (good teaching with appropriate intermediate challenge)"
+      },
+      "advanced": {
+        "focus": "Deep principles, minimal scaffolding (Level 1-2), nuanced edge cases, creative transfer",
+        "diagnostic": "Probe for subtle misconceptions, test boundary understanding, assess meta-cognitive awareness",
+        "scaffolding": "Start Level 2-3 (independent with feedback), fade to transfer (teach others), mostly questions",
+        "validation": "Far transfer and creative synthesis, teaching test, Bloom Level 5-6 (evaluate, create)",
+        "typical_score": "Aim for 4.5-5.0 (excellent teaching with deep validation and transfer)"
+      }
+    },
+    "by_topic_complexity": {
+      "concrete_procedural": {
+        "examples": "Tool usage, algorithm implementation, step-by-step procedures",
+        "question_focus": "Code tracing, step-by-step walkthrough, 'What happens next?'",
+        "scaffolding": "Worked examples → partial examples → independent practice",
+        "validation": "Near transfer (similar procedures), can perform without hints",
+        "common_pitfalls": "Over-scaffolding (doesn't fade), teaching memorization not understanding"
+      },
+      "abstract_conceptual": {
+        "examples": "Recursion, object-oriented design, quantum mechanics, philosophical concepts",
+        "question_focus": "Concrete analogies first, pattern recognition, formalization, edge cases",
+        "scaffolding": "Build from concrete to abstract, use multiple analogies, think-aloud modeling",
+        "validation": "Far transfer (recognize pattern in new domain), multi-level explanations (ELI5 to expert)",
+        "common_pitfalls": "Starting too abstract, single analogy (doesn't generalize), skipping concrete foundation"
+      },
+      "misconception_heavy": {
+        "examples": "Statistics (correlation vs causation), physics (force vs acceleration), programming (references vs values)",
+        "question_focus": "Prediction-observation-explanation, multiple counterexamples, 'Why does [surprising case] happen?'",
+        "scaffolding": "Build correct model from scratch (don't just attack wrong model), spaced retrieval checks",
+        "validation": "Misconception elimination verified with novel test cases, persistence checks (spaced retrieval)",
+        "common_pitfalls": "Single correction (misconception returns), assertion instead of discovery, no persistence check"
+      }
+    }
+  },
+  "common_failure_modes": {
+    "guessing_game": {
+      "symptom": "Teacher asks questions until learner guesses 'correct' answer, then moves on. Feels like 'What am I thinking?' rather than genuine discovery.",
+      "root_cause": "Teacher has specific answer in mind, doesn't honor learner's reasoning process, rushes to confirmation",
+      "fix": "Ask 'How would you figure this out?' not 'What's the answer?' Honor alternative reasoning paths. If learner's logic is sound but conclusion differs, explore their path. Provide thinking time (30 seconds minimum)."
+    },
+    "no_scaffolding_adjustment": {
+      "symptom": "Same level of support throughout session regardless of learner success or struggle. Either constant full scaffolding or constant independence.",
+      "root_cause": "Not monitoring learner signals (boredom, productive struggle, frustration). One-size-fits-all approach.",
+      "fix": "Explicitly track scaffolding level per concept. Success signal → fade down one level. Struggle >2min → increase one level. Monitor for ZPD signals (too easy, optimal, too hard)."
+    },
+    "misconception_assertion": {
+      "symptom": "When misconception appears, teacher says 'No, it's actually X' without exploring why learner thought otherwise.",
+      "root_cause": "Treating misconception as simple error rather than coherent (but incorrect) mental model",
+      "fix": "Use prediction-observation-explanation protocol: Get explicit prediction based on misconception → Show contradiction → Guide discovery of correct model through questions. Verify with new examples and spaced retrieval."
+    },
+    "no_transfer_validation": {
+      "symptom": "Session ends when learner can repeat back explanation or solve identical problem. No testing with novel cases.",
+      "root_cause": "Mistaking procedural recall for understanding. Not testing if learner can transfer to new contexts.",
+      "fix": "Always test transfer: Near (similar problem), Far (analogous structure in different domain). Feynman test (explain at multiple levels). Teaching test ('How would you teach this? What would be confusing?')."
+    },
+    "starting_too_abstract": {
+      "symptom": "Jumps to formal definitions and technical terms before building intuitive understanding. Learner nods but doesn't truly get it.",
+      "root_cause": "Expertise blind spot—teacher forgets what's hard for beginners. Assumes intuitions that need to be built.",
+      "fix": "Always start concrete (real-world analogy, simple example). Build pattern recognition. Only then formalize with terminology. Progression: Concrete → Pattern → Formal → Edge Cases → Transfer."
+    },
+    "insufficient_diagnostic": {
+      "symptom": "Starts teaching without assessing learner's current understanding. Wastes time on things they already know or assumes knowledge they don't have.",
+      "root_cause": "Impatience to start 'teaching'. Assumes baseline knowledge level.",
+      "fix": "Always start with 3-5 diagnostic questions: What do you know about X? Can you give example of Y? What do you think Z means? How would you approach [problem]? Use answers to identify gaps and misconceptions."
+    },
+    "persistent_misconception_ignored": {
+      "symptom": "Misconception corrected once, seems fixed, but returns later when learner is under time pressure or in new context.",
+      "root_cause": "Surface compliance (learner says 'right' answer) vs deep mental model change. Single correction insufficient for deep misconceptions.",
+      "fix": "Spaced retrieval checks (test same misconception days later in different context). For deep misconceptions: build correct model from scratch, provide 3-5 diverse counterexamples (not dismissible as 'special cases'), use role reversal ('How would you correct someone with this misconception?')."
+    },
+    "no_next_steps": {
+      "symptom": "Session ends abruptly after learner 'gets it'. No guidance on how to solidify learning or what to practice.",
+      "root_cause": "Treating session as one-off rather than part of learning journey. No learning continuity planning.",
+      "fix": "End every session with: (1) Summary of key insights, (2) Specific practice recommendations (problem types, not just 'practice more'), (3) Spaced retrieval schedule (review in 1, 3, 7 days), (4) Success criteria for mastery ('You'll know you've mastered this when...')."
+    }
+  },
+  "excellence_indicators": [
+    "Diagnostic reveals specific misconceptions with evidence (not assumptions)",
+    "Question ladder has clear structure (concrete → pattern → formal → edge → transfer)",
+    "Every question has explicit purpose (not Socratic theater or guessing game)",
+    "Scaffolding starts at appropriate level (based on diagnostic, not defaulting to full modeling)",
+    "Scaffolding fades systematically as competence grows (tracks per concept)",
+    "Misconceptions corrected through contradiction and discovery (not assertion)",
+    "Multiple counterexamples for deep misconceptions (not dismissible as special cases)",
+    "Spaced retrieval checks for persistent misconceptions (verification, not single correction)",
+    "ZPD maintained (learner engaged in productive struggle, neither bored nor frustrated)",
+    "Transfer validated at multiple levels (near, far, creative synthesis)",
+    "Feynman test passed (can explain at child, peer, expert, and teacher levels)",
+    "Teaching test passed ('How would you teach this? What would confuse learners?')",
+    "All identified misconceptions verified eliminated with novel test cases",
+    "Honors learner's alternative reasoning paths (not just fishing for teacher's answer)",
+    "Provides learning continuity (summary, practice plan, spaced review, success criteria)",
+    "Session would produce durable understanding (not just procedural recall)"
+  ]
+}

skills/socratic-teaching-scaffolds/resources/methodology.md

@@ -0,0 +1,468 @@
+# Advanced Socratic Teaching Methodology
+
+## Workflow
+
+Copy this checklist for complex teaching scenarios:
+
+```
+Advanced Teaching Progress:
+- [ ] Step 1: Deep diagnostic with misconception mapping
+- [ ] Step 2: Multi-ladder design for complex topics
+- [ ] Step 3: Adaptive questioning with branching
+- [ ] Step 4: Strategic scaffolding fading
+- [ ] Step 5: Deep transfer validation
+```
+
+**Step 1: Deep diagnostic** - Use advanced probing to map mental models and nested misconceptions. See [1. Advanced Diagnostic Techniques](#1-advanced-diagnostic-techniques).
+
+**Step 2: Multi-ladder design** - Build parallel question sequences for multi-faceted concepts. See [2. Multi-Ladder Design](#2-multi-ladder-design).
+
+**Step 3: Adaptive questioning** - Branch based on learner responses, handle persistent misconceptions. See [3. Adaptive Questioning](#3-adaptive-questioning).
+
+**Step 4: Strategic scaffolding** - Use advanced fading patterns and apprenticeship models. See [4. Strategic Scaffolding Fading](#4-strategic-scaffolding-fading).
+
+**Step 5: Deep transfer** - Validate understanding across multiple abstraction levels and domains. See [5. Deep Transfer Validation](#5-deep-transfer-validation).
+
+---
+
+## 1. Advanced Diagnostic Techniques
+
+### Mental Model Elicitation
+
+**Technique: Concept Mapping Interview**
+- "Draw/describe how [concepts] relate to each other"
+- Look for: Missing connections, incorrect causal arrows, confused hierarchies
+- Example: Teaching recursion → Ask them to draw relationship between function, call stack, base case
+
+**Technique: Predict-Observe-Explain (POE)**
+- Present scenario: "What will happen when [test case]?"
+- Observe their prediction (reveals mental model)
+- Show actual outcome
+- Ask: "Why different from prediction?"
+
+**Technique: Analogical Reasoning Probe**
+- "This is like [analogy]. How is it similar? How is it different?"
+- Mismatched analogies reveal misconceptions
+- Example: "Is recursion like a loop?" (reveals whether they understand call stack vs iteration)
+
+### Misconception Taxonomy
+
+**Surface vs Deep Misconceptions:**
+
+**Surface** (Easy to fix with single correction):
+- Terminology confusion ("pointer" vs "reference")
+- Memorization errors (wrong formula)
+- Single faulty assumption
+
+**Deep** (Require rebuilding mental model):
+- Fundamental misunderstanding (thinking correlation implies causation)
+- Coherent but wrong model (Aristotelian physics: heavier objects fall faster)
+- Transferred wrong pattern (applying linear thinking to exponential problems)
+
+**Diagnostic Questions by Type:**
+
+| Misconception Type | Question to Reveal | Correct Understanding |
+|-------------------|-------------------|----------------------|
+| Causal reversal | "Does A cause B or B cause A?" | Identify correct direction |
+| False dichotomy | "Is it X or Y?" (when both/neither) | Reveal multiple possibilities |
+| Overgeneralization | "Does this always hold?" | Show edge cases/boundaries |
+| Undergeneralization | "When else would this apply?" | Extend to broader contexts |
+| Confused levels | "Is this about [high level] or [low level]?" | Separate abstraction layers |
+
+### Prior Knowledge Mapping
+
+**Backward Chaining from Target:**
+1. What must they know before understanding target concept?
+2. What must they know before that?
+3. Continue until you reach confirmed knowledge
+
+**Example (Teaching Big-O Notation):**
+- Target: Understand O(n²) vs O(n log n)
+- Prerequisite: Understand growth rates
+- Prerequisite: Understand functions
+- Prerequisite: Understand variables
+- Start teaching at first gap
+
+**Knowledge Dependency Graph:**
+```
+Target Concept
+├── Prerequisite A
+│   ├── Sub-prerequisite A1
+│   └── Sub-prerequisite A2
+├── Prerequisite B
+└── Prerequisite C (MISSING ← Start here)
+```
+
+---
+
+## 2. Multi-Ladder Design
+
+For complex topics requiring multiple complementary question sequences:
+
+### Parallel Ladders Strategy
+
+**When to use:** Topic has multiple independent facets that all need understanding
+
+**Example: Teaching Object-Oriented Programming**
+
+**Ladder 1: Encapsulation**
+1. Why hide data inside object?
+2. What happens if everything is public?
+3. How do getters/setters help?
+
+**Ladder 2: Inheritance**
+1. What code would we duplicate without inheritance?
+2. Is-a vs has-a relationships?
+3. When does inheritance hurt?
+
+**Ladder 3: Polymorphism**
+1. How to treat different objects uniformly?
+2. What's the interface contract?
+3. Static vs dynamic dispatch?
+
+**Integration Point:**
+"How do these three ideas work together in [system design problem]?"
+
+### Spiral Curriculum Approach
+
+**Pattern:** Revisit concept at increasing depth levels across multiple sessions
+
+**Session 1 (Intuition):** Concrete examples, basic mental model
+**Session 2 (Application):** Use in simple problems, edge cases
+**Session 3 (Formalization):** Technical terminology, precise definitions
+**Session 4 (Transfer):** Apply to novel domains, teach others
+
+**Advantage:** Each pass deepens understanding without overwhelming
+
+### Concept Lattice Navigation
+
+**Structure:** Concepts form lattice (partial order) not linear sequence
+
+```
+        Abstract Concept
+       /                \
+    Aspect A          Aspect B
+       |                  |
+    Example A1        Example B1
+```
+
+**Navigation strategies:**
+- **Breadth-first:** Cover all aspects at high level, then drill down
+- **Depth-first:** Master one aspect completely, then move to next
+- **Learner-directed:** "Want to go deeper here, or explore different angle?"
+
+---
+
+## 3. Adaptive Questioning
+
+### Branching Question Trees
+
+**Structure:**
+
+```
+Q1: Diagnostic question
+├─ Correct → Q2A (advance)
+├─ Misconception M1 → Q2B (address M1) → Q2C (verify correction) → Q2A
+└─ Stuck → Scaffold → Q1 (retry)
+```
+
+**Implementation:**
+- Prepare 2-3 follow-up paths for each question
+- Common misconception → Specific correction sequence
+- Stuck → Scaffolding question → Return to original
+- Correct → Advance to next level
+
+### Misconception-Specific Interventions
+
+**For Persistent Misconceptions:**
+
+**Technique 1: Multiple Contradictions**
+- Single counterexample often dismissed as "special case"
+- Provide 3-5 diverse counterexamples
+- Ask: "What do all these have in common?"
+
+**Technique 2: Extreme Cases**
+- Push misconception to absurd conclusion
+- "If that were true, what would happen when [extreme]?"
+- Learner recognizes absurdity → reconsiders
+
+**Technique 3: Role Reversal**
+- "You're the teacher. Student says [misconception]. How would you correct them?"
+- Explaining to others often clarifies own thinking
+
+**Technique 4: Historical Misconception**
+- "Many scientists thought [misconception] until [discovery]. Why did they think that? What changed?"
+- Legitimizes struggle, shows path to correct understanding
+
+### Responsive Scaffolding Triggers
+
+**Student Signal** → **Scaffolding Response**
+
+| Signal | What It Means | Appropriate Response |
+|--------|---------------|---------------------|
+| Silent >30s, engaged | Productive struggle | Wait, don't interrupt |
+| Silent >2min, disengaged | Stuck/frustrated | Provide hint or scaffolding |
+| Partially correct answer | Close, minor gap | "Almost! What about [aspect]?" |
+| Confident wrong answer | Misconception | POE: predict outcome, show contradiction |
+| Multiple failed attempts | Too large leap | Break into smaller steps |
+| "I don't know where to start" | Missing entry point | Provide concrete example to anchor |
+
+---
+
+## 4. Strategic Scaffolding Fading
+
+### Cognitive Apprenticeship Model
+
+**Phase 1: Modeling** (Teacher demonstrates with thinking aloud)
+- "Watch how I approach this problem..."
+- Articulate every decision: "I'm choosing X because Y"
+- Make invisible thinking visible
+
+**Phase 2: Coaching** (Student attempts, teacher guides)
+- "Try it. I'll watch and give hints."
+- Intervene before errors compound
+- Ask guiding questions, don't give answers
+
+**Phase 3: Scaffolding** (Teacher provides structure, student fills in)
+- "I'll set up the problem. You solve it."
+- "Here's the outline. Add the details."
+- Temporary support, explicitly temporary
+
+**Phase 4: Articulation** (Student explains their thinking)
+- "Walk me through your reasoning."
+- "Why did you choose that approach?"
+- Makes their thinking explicit to themselves
+
+**Phase 5: Reflection** (Compare approaches, identify strategies)
+- "How does your solution compare to mine?"
+- "When would your approach work better?"
+- Meta-cognitive awareness
+
+**Phase 6: Exploration** (Student tackles novel problems independently)
+- "Here's a related but different problem. Try it."
+- No scaffolding unless requested
+- Transfer to new contexts
+
+### Fading Dimensions
+
+**Fade Multiple Aspects Separately:**
+
+**Dimension 1: Problem Complexity**
+- Start: Single-step problems
+- Middle: Multi-step with clear path
+- End: Multi-step with multiple viable paths
+
+**Dimension 2: Hints Provided**
+- Start: Explicit hints at each step
+- Middle: Hints only when stuck
+- End: No hints, only verification
+
+**Dimension 3: Example Completeness**
+- Start: Fully worked example
+- Middle: Partial example (starter code)
+- End: No example, just specification
+
+**Strategy:** Fade one dimension at a time to avoid overwhelming
+
+### Zone of Proximal Development (ZPD) Calibration
+
+**Too Easy (Below ZPD):**
+- Symptoms: Boredom, quick correct answers without thought
+- Adjustment: Skip ahead, increase complexity
+
+**Optimal (Within ZPD):**
+- Symptoms: Engaged struggle, eventual success with hints
+- Maintain: Current scaffolding level
+
+**Too Hard (Above ZPD):**
+- Symptoms: Frustration, wild guesses, giving up
+- Adjustment: Increase scaffolding, break into smaller steps
+
+**Dynamic Adjustment:**
+- Start conservative (more scaffolding)
+- Fade aggressively when success
+- Reinstate scaffolding immediately when struggle turns to frustration
+
+---
+
+## 5. Deep Transfer Validation
+
+### Transfer Assessment Taxonomy
+
+**Level 1: Near Transfer (Same domain, similar problem)**
+- Given: Taught quicksort
+- Test: "Sort this different array using quicksort"
+- Validates: Procedural memory
+
+**Level 2: Modified Transfer (Same domain, modified problem)**
+- Given: Taught quicksort
+- Test: "Adapt quicksort to find kth smallest element"
+- Validates: Flexible application
+
+**Level 3: Far Transfer (Different domain, analogous structure)**
+- Given: Taught quicksort (divide-and-conquer)
+- Test: "Use divide-and-conquer to solve [unrelated problem]"
+- Validates: Deep principle extraction
+
+**Level 4: Creative Transfer (Novel synthesis)**
+- Given: Taught multiple algorithms
+- Test: "Design new algorithm for [novel problem]"
+- Validates: Generative understanding
+
+### Feynman Understanding Test
+
+**Depth Levels:**
+
+**Level 1: Explanation to Child (ELI5)**
+- No technical jargon
+- Simple analogies
+- Tests: Can they find intuitive core?
+
+**Level 2: Explanation to Peer**
+- Some terminology
+- Concrete examples
+- Tests: Can they make it relatable?
+
+**Level 3: Explanation to Expert**
+- Technical precision
+- Edge cases and limitations
+- Tests: Can they be rigorous?
+
+**Level 4: Teaching While Handling Misconceptions**
+- Anticipate confusions
+- Prepare counterexamples
+- Tests: Meta-cognitive understanding of learning process
+
+**Assessment:** True understanding = Can explain at all levels
+
+### Bloom's Taxonomy Validation
+
+**Level 1: Remember**
+- "What is [definition]?"
+- Tests: Recall only
+
+**Level 2: Understand**
+- "Explain [concept] in your own words"
+- Tests: Comprehension
+
+**Level 3: Apply**
+- "Use [concept] to solve [problem]"
+- Tests: Procedural knowledge
+
+**Level 4: Analyze**
+- "Why does [approach] work for [case] but fail for [other case]?"
+- Tests: Principled understanding
+
+**Level 5: Evaluate**
+- "Which solution is better and why?"
+- Tests: Critical judgment
+
+**Level 6: Create**
+- "Design a [new thing] using [concept]"
+- Tests: Generative mastery
+
+**Teaching Target:** Aim for Levels 3-4 minimum, 5-6 for mastery
+
+---
+
+## 6. Domain-Specific Patterns
+
+**Programming:** Code tracing ("What does this do?" → "Trace with input X" → "Why?"), debugging buggy code, refactoring exercises
+
+**Math/Science:** Proof discovery ("Find counterexample or prove"), dimensional analysis (unit checking), limiting cases (parameter → 0 or ∞)
+
+**Conceptual:** Thought experiments (trolley problem, Schrödinger's cat → "What would you do?" → "Why?"), Socratic dialogue (probe assumptions until contradiction)
+
+---
+
+## 7. Persistent Misconception Strategies
+
+### Common Failure Modes & Fixes
+
+**Problem: Misconception Returns After Seeming Correction**
+
+**Cause:** Surface compliance vs deep understanding
+- Learner says "correct" answer but hasn't changed mental model
+- Under time pressure, reverts to misconception
+
+**Fix:** Spaced retrieval
+- Test understanding days later
+- Ask same question in different context
+- Multiple spaced exposures required
+
+**Problem: Learner Stuck in Wrong Model**
+
+**Cause:** Current model is coherent and explains many phenomena
+- Example: Aristotelian physics (heavier falls faster - explains cannonball vs feather in air)
+
+**Fix:** Build correct model from scratch before dismantling wrong one
+- Don't just show counterexamples
+- Construct alternative explanation
+- Then show new model explains everything old model did PLUS counterexamples
+
+**Problem: Guessing Instead of Reasoning**
+
+**Cause:** Fishing for "correct answer" instead of thinking
+
+**Fix:** Make process more important than answer
+- "Don't tell me the answer. Tell me how you'd figure it out."
+- "Even if wrong, explain your reasoning."
+- Reward process, not just correct answers
+
+### Misconception Resistance Hierarchy
+
+**Level 1: Fragile** (Single correction sufficient)
+- Example: Wrong terminology
+- Fix: Correct and provide correct term
+
+**Level 2: Moderate** (Need 2-3 corrections in different contexts)
+- Example: Confused variable scope
+- Fix: Show scope behavior in multiple code examples
+
+**Level 3: Robust** (Requires rebuilding mental model)
+- Example: Thinking objects are copied by default in Python
+- Fix: Explain reference semantics from scratch, trace through multiple examples
+
+**Level 4: Foundational** (Requires prerequisite knowledge first)
+- Example: Understanding quantum mechanics while thinking deterministically
+- Fix: First teach probability/statistics, THEN quantum
+
+**Strategy:** Identify resistance level, apply appropriate intervention intensity
+
+---
+
+## 8. Self-Directed Learning Design
+
+**Self-Paced Module Structure:** Pre-assessment (can you already?) → Learning objective → Worked example → Guided practice (partial examples + hints) → Independent practice → Self-check with explanations
+
+**Hint System:** Hidden by default, progressive revelation (3-5 hints from gentle to explicit), last "hint" is full solution
+
+**Question Types:** Recall (definitions), application (solve problems), analysis (why it works), misconception checks (T/F common errors)
+
+**Rich Feedback:** Not just correct/incorrect. Wrong → "This suggests [misconception]. Actually, [correction]." Correct → "Right because [principle]."
+
+**Spaced Repetition:** Review at 1, 3, 7, 14 days, then monthly
+
+---
+
+## 9. Quality Indicators
+
+**Excellent Socratic Teaching:**
+- [ ] Learner discovers insights themselves (not told)
+- [ ] Questions reveal thinking (not guess teacher's answer)
+- [ ] Scaffolding fades as competence grows
+- [ ] Misconceptions corrected through contradiction, not assertion
+- [ ] Can explain concept at multiple levels (ELI5 → Expert)
+- [ ] Transfers to novel problems without prompting
+- [ ] Asks good questions themselves (meta-cognitive growth)
+
+**Poor Pseudo-Socratic Teaching:**
+- [ ] Questions are just a guessing game
+- [ ] Teacher gives answer when learner doesn't guess "correctly"
+- [ ] No scaffolding adjustment (one-size-fits-all)
+- [ ] Misconceptions ignored or corrected by fiat
+- [ ] Only one explanation level (usually too technical)
+- [ ] Can only solve problems identical to examples
+- [ ] Passive consumption, no active discovery
+
+**Assessment:** More checks in "Excellent" → Teaching is effective

skills/socratic-teaching-scaffolds/resources/template.md

@@ -0,0 +1,366 @@
+# Socratic Teaching Session Template
+
+## Workflow
+
+Copy this checklist and track your progress:
+
+```
+Teaching Session Progress:
+- [ ] Step 1: Diagnose current understanding
+- [ ] Step 2: Build question ladder
+- [ ] Step 3: Execute teaching session
+- [ ] Step 4: Fade scaffolding
+- [ ] Step 5: Validate understanding
+```
+
+**Step 1: Diagnose current understanding** - Ask probing questions to assess baseline knowledge and misconceptions. See [Section 1](#1-diagnostic-phase).
+
+**Step 2: Build question ladder** - Design progression from current to target understanding. See [Section 2](#2-question-ladder-design).
+
+**Step 3: Execute teaching session** - Guide discovery through questions and scaffolding. See [Section 3](#3-teaching-session-structure).
+
+**Step 4: Fade scaffolding** - Progressively remove support as competence grows. See [Section 4](#4-scaffolding-fading-protocol).
+
+**Step 5: Validate understanding** - Test transfer and misconception elimination. See [Section 5](#5-validation--assessment).
+
+---
+
+## 1. Diagnostic Phase
+
+### Learning Profile
+
+**Learner Information:**
+- **Name/Role**: [Who is learning]
+- **Goal**: [What they want to achieve]
+- **Timeline**: [When they need to know it]
+- **Current Experience**: [Novice / Beginner / Intermediate / Advanced in this domain]
+
+**Topic to Teach:**
+- **Concept/Skill**: [Specific topic]
+- **Why Important**: [Motivation, application context]
+- **Success Criteria**: [How will we know they've learned it?]
+
+### Diagnostic Questions
+
+Ask 3-5 questions to assess current understanding:
+
+1. **Clarifying**: "What do you already know about [topic]?"
+2. **Probing**: "Can you give me an example of [related concept]?"
+3. **Assumption**: "What do you think [key term] means?"
+4. **Application**: "How would you approach [simple problem]?"
+5. **Misconception Check**: "[Question that reveals common misconception]"
+
+**Identified Knowledge Gaps:**
+- [ ] Gap 1: [What they don't know yet]
+- [ ] Gap 2: [What they don't know yet]
+- [ ] Gap 3: [What they don't know yet]
+
+**Identified Misconceptions:**
+- [ ] Misconception 1: [Faulty mental model detected]
+- [ ] Misconception 2: [Faulty mental model detected]
+
+**Starting Scaffolding Level:** [Full Modeling / Guided Practice / Coached Practice / Independent]
+
+---
+
+## 2. Question Ladder Design
+
+Build progression from current understanding to target concept:
+
+### Concrete Foundation (Step 1-2)
+
+**Analogy/Real-World Example:**
+- Anchor concept in familiar experience
+- Example: [Concrete situation they already understand]
+
+**Questions:**
+1. [Simple question connecting to existing knowledge]
+2. [Question exploring the analogy]
+
+### Pattern Recognition (Step 3-4)
+
+**Key Pattern to Notice:**
+- What regularities or structures should they see?
+- Pattern: [Core insight they should discover]
+
+**Questions:**
+3. [Question that guides to pattern]
+4. [Question that reinforces pattern with new example]
+
+### Formalization (Step 5-6)
+
+**Technical Vocabulary:**
+- Introduce precise terminology once pattern is clear
+- Terms: [Key terms to define]
+
+**Questions:**
+5. [Question that motivates formal definition]
+6. [Question applying formal concept]
+
+### Edge Cases & Boundaries (Step 7-8)
+
+**Limitations to Explore:**
+- Where does the concept break down or need qualification?
+- Edge cases: [Boundary conditions]
+
+**Questions:**
+7. [Question revealing edge case]
+8. [Question exploring why edge case matters]
+
+### Transfer & Application (Step 9-10)
+
+**Novel Context:**
+- Apply concept to new situation
+- Context: [Different domain or problem]
+
+**Questions:**
+9. [Question requiring transfer to novel situation]
+10. [Question checking deep understanding]
+
+---
+
+## 3. Teaching Session Structure
+
+### Opening (5 minutes)
+
+**State Goal:**
+"Today we're going to understand [concept]. By the end, you'll be able to [success criteria]."
+
+**Check Motivation:**
+"Why is this important to you?" [Learner answers]
+
+### Main Teaching Loop (30-45 minutes)
+
+For each question in ladder:
+
+**Ask Question:**
+- State question clearly
+- Give thinking time (30 seconds minimum)
+- Don't rush to hint
+
+**Observe Response:**
+- **If correct understanding**: Confirm and move to next question
+- **If partial understanding**: Ask follow-up to clarify
+- **If misconception revealed**: Note it, explore contradiction (see [Misconception Protocol](#misconception-correction-protocol))
+- **If stuck**: Provide scaffolding (see [Scaffolding Menu](#scaffolding-menu))
+
+**Scaffold if Needed:**
+- Level 5 (Modeling): "Let me show you how I'd think about this..."
+- Level 4 (Guided): "What if we try [partial solution]?"
+- Level 3 (Coached): "You're close. What about [hint]?"
+- Level 2 (Independent): "Take your time. Walk me through your thinking."
+
+**Check Understanding:**
+- "Can you explain that in your own words?"
+- "How does that connect to [earlier concept]?"
+
+### Closing (10 minutes)
+
+**Summary:**
+"Let's review what we covered: [key points]"
+
+**Transfer Task:**
+"Now try [novel problem using concept]"
+
+**Next Steps:**
+"To solidify this, [practice recommendation]"
+
+---
+
+## 4. Scaffolding Fading Protocol
+
+Track scaffolding level for each concept:
+
+| Concept/Skill | Initial Level | Current Level | Target: Independent |
+|---------------|---------------|---------------|---------------------|
+| [Concept 1]   | Level 5       | Level 4       | [ ] Ready           |
+| [Concept 2]   | Level 4       | Level 3       | [ ] Ready           |
+| [Concept 3]   | Level 3       | Level 2       | [ ] Ready           |
+
+**Fading Triggers:**
+- **Success Signal**: Learner completes task correctly → Move down one level
+- **Struggle Signal**: Learner stuck for >2 minutes → Move up one level
+- **Frustration Signal**: Repeated failure or negative emotion → Provide direct explanation, restart with higher level
+
+**Progressive Independence:**
+1. Start: Full worked example with narration
+2. Next: Partial example, learner completes
+3. Next: Learner attempts, teacher provides hints
+4. Next: Learner attempts, teacher reviews after
+5. End: Learner explains concept to someone else
+
+---
+
+## 5. Validation & Assessment
+
+### Understanding Checks
+
+**Explanation Test:**
+- "Explain [concept] to me like I'm [5 years old / your colleague / an expert]"
+- Quality: Clear, accurate, appropriate detail for audience
+
+**Application Test:**
+- "Use [concept] to solve [novel problem]"
+- Quality: Correct application, adapts to new context
+
+**Teaching Test:**
+- "How would you teach this to someone else?"
+- Quality: Can identify key questions, common misconceptions
+
+### Misconception Elimination
+
+Check that identified misconceptions are corrected:
+
+- [ ] Misconception 1: **Eliminated?** [Test question] → [Response shows correction]
+- [ ] Misconception 2: **Eliminated?** [Test question] → [Response shows correction]
+
+**If misconception persists:**
+- Design new question sequence targeting it specifically
+- See [resources/methodology.md](../methodology.md) for advanced misconception busting
+
+### Transfer Assessment
+
+**Near Transfer** (same domain, different problem):
+- Problem: [Similar but not identical]
+- Success: [ ] Solved correctly without hints
+
+**Far Transfer** (different domain, analogous structure):
+- Problem: [Different context, same underlying principle]
+- Success: [ ] Recognized analogous structure and applied concept
+
+---
+
+## Scaffolding Menu
+
+Quick reference for providing appropriate support:
+
+**Level 5: Full Modeling**
+- "Let me show you a complete example..."
+- "Here's how I would think through this step-by-step..."
+- "Watch how I approach [problem], then you'll try a similar one"
+
+**Level 4: Guided Practice**
+- "I'll start, and you complete the next steps..."
+- "Here's the first part [show partial solution]. Can you finish?"
+- "Let's do this together. I'll guide you through each step."
+
+**Level 3: Coached Practice**
+- "Give it a try. I'll ask questions if you get stuck."
+- "You're on the right track. What about [specific aspect]?"
+- "Almost. Think about what would happen if [scenario]?"
+
+**Level 2: Independent with Feedback**
+- "Try it yourself first. We'll review together afterwards."
+- "Take your time. Come back when you have a solution."
+- "Work through this, then explain your reasoning."
+
+**Level 1: Transfer**
+- "Now teach this to [someone else]."
+- "Create your own example problem."
+- "Explain why someone might misunderstand this."
+
+---
+
+## Misconception Correction Protocol
+
+When misconception is revealed:
+
+**Step 1: Acknowledge Without Judgment**
+- "Interesting! Many people think that."
+- "That's a really common way to think about it."
+
+**Step 2: Predict Outcome Based on Misconception**
+- "If [misconception] were true, what would we expect to see in [test case]?"
+- Get learner to make explicit prediction
+
+**Step 3: Show Contradiction**
+- Demonstrate or explain actual outcome
+- "But actually, [show real result]. Why do you think that is?"
+
+**Step 4: Guide to Correct Model**
+- Ask questions that lead to correct understanding
+- "What could explain this difference?"
+- Don't just state correct answer—guide discovery
+
+**Step 5: Reinforce with New Examples**
+- Apply corrected understanding to 2-3 new cases
+- "Let's test this new understanding. What about [example]?"
+
+**Step 6: Check Persistence**
+- Return to misconception trigger later in session
+- Ensure correction stuck, not just surface compliance
+
+---
+
+## Common Teaching Patterns
+
+**Pattern: Concrete → Abstract (Feynman Technique)**
+
+1. **Level 1 (Child)**: Simple analogy, no jargon
+   - "Think of it like [everyday object/experience]..."
+
+2. **Level 2 (High School)**: Introduce some formality
+   - "More precisely, it's when [definition with some technical terms]..."
+
+3. **Level 3 (Undergraduate)**: Full technical definition
+   - "Formally, [concept] is defined as [precise definition with terminology]..."
+
+4. **Level 4 (Graduate)**: Edge cases, formal proofs
+   - "Under these conditions [constraints], we can prove [property]..."
+
+**Pattern: Problem → Decomposition → Solution**
+
+1. **Present Complex Problem**: Something they can't solve yet
+2. **Ask Decomposition Questions**: "What are the sub-problems?"
+3. **Solve Simple Sub-Problem**: Build confidence with achievable piece
+4. **Compose**: "How do we combine these solutions?"
+5. **Generalize**: "What pattern did we use? When else could we apply it?"
+
+**Pattern: Prediction → Observation → Explanation**
+
+1. **Predict**: "What do you think will happen if [scenario]?"
+2. **Observe**: [Show actual outcome—contradicts naive prediction]
+3. **Explain**: "Why was our prediction wrong? What's really happening?"
+4. **Refine Model**: "Let's adjust our understanding to account for this..."
+
+---
+
+## Quality Checklist
+
+Before concluding session, verify:
+
+**Diagnostic:**
+- [ ] Asked 3-5 diagnostic questions to assess baseline
+- [ ] Identified specific knowledge gaps
+- [ ] Detected at least 1 misconception (if present)
+- [ ] Determined appropriate starting scaffolding level
+
+**Question Ladder:**
+- [ ] Built progression from concrete to abstract (min 8 questions)
+- [ ] Each question has clear purpose (not just Socratic theater)
+- [ ] Ladder addresses identified gaps and misconceptions
+- [ ] Questions build on each other logically
+
+**Teaching Execution:**
+- [ ] Started at appropriate scaffolding level (not always Level 5)
+- [ ] Faded scaffolding as competence increased
+- [ ] Asked questions, didn't just lecture
+- [ ] Corrected misconceptions through contradiction, not assertion
+- [ ] Adjusted to learner responses (didn't stick rigidly to script)
+
+**Validation:**
+- [ ] Tested understanding with novel problem (transfer)
+- [ ] Asked for explanation in learner's words
+- [ ] Verified misconceptions eliminated
+- [ ] Provided next steps for continued learning
+
+**Guardrails:**
+- [ ] Stayed in zone of proximal development (optimal challenge)
+- [ ] Didn't make it a guessing game
+- [ ] Made implicit knowledge explicit
+- [ ] Adapted to learner's pace and preferences
+
+**Session Notes:**
+- What worked well: [Note effective moments]
+- What to adjust: [Note what to improve]
+- Follow-up needed: [Topics requiring more work]