| name | mental-model-toolkit |
| description | Use when someone is stuck on a problem and needs a different frame entirely, suspects they have blind spots, or has exhausted obvious solutions. Three instruments: choose the right mental model, systematically surface unknown unknowns, and break... |
| scenarios | ["I've tried everything obvious — what am I missing?","Help me find blind spots in this plan","Give me a completely different frame for this problem","뻔한 해결책 말고 다른 관점으로 접근해줘","이 계획에서 내가 놓치고 있는 게 뭘까?","이 문제를 다른 방식으로 볼 수 있어?"] |
| compatibility | {"recommended":["think-tool"],"optional":["sequential-thinking","mcp-reasoner"],"remote_mcp_note":"think-tool이 있으면 어떤 도구(mental model·unknown unknowns·lateral thinking)가 이 문제에 맞는지 정확히 진단할 수 있습니다. mcp-reasoner는 lateral thinking 단계에서 다중 경로 탐색에 유용합니다. Claude 설정 → MCP Servers에서 remote SSE 엔드포인트를 추가하세요."} |
Mental Model Toolkit
When to Use / When Not to Use
Use when:
- All obvious solutions have been tried and the problem persists
- A plan is being finalized and needs stress-testing for hidden risks
- A fresh frame would unlock options not visible from the current angle
Not for:
- Mapping downstream consequences (use second-order-thinker)
- Auditing specific assumptions (use assumption-extractor)
Process
Diagnose which instrument fits before applying:
| Situation | Instrument |
|---|
| Standard problem, want rigorous framing | Mental model library |
| Confident in plan, want stress-test | Unknown unknowns mapper |
| Stuck, obvious solutions tried | Lateral thinking prompter |
| Surprised by outcome, diagnosing cause | Map vs territory + perspective grid |
Instrument 1 — Mental Model Library
Key models: First Principles Thinking, Inversion, Map vs Territory, Occam's Razor, Second-Order Thinking, Pareto (80/20), Regret Minimization, Chesterton's Fence.
Full catalog: see references/mental-models-catalog.md
Apply the model that fits the problem — not the one you already know best. Use the model to actually derive something, not as decoration.
Instrument 2 — Unknown Unknowns Mapper
You cannot observe blind spots directly. Create conditions where they become visible.
- Pre-mortem: Imagine it's 12 months out and this failed spectacularly. What happened? Generate 5+ failure modes.
- Red teaming: Argue the case against the plan without defending it.
- Outsider perspective: Explain to someone unfamiliar — their confusion points to unexamined assumptions.
- Assumption audit: List every assumption; rate confidence (0–100%) and impact if wrong.
- Perspective expansion: Inhabit each viewpoint (user, skeptic, future self, opposite worldview, most harmed).
Instrument 3 — Lateral Thinking Prompter
Techniques: Random Entry, Provocation (Po), Reversal, Analogical Thinking, Constraint Removal.
Full reference: see references/lateral-thinking-techniques.md
MCP note: If mcp-reasoner is available, use it here — lateral thinking requires multi-path exploration before selecting the best creative approach.
Output Template
- Diagnosis — which instrument fits and why
- Application — work through the chosen technique with the specific problem
- Outputs — new frames, blind spots surfaced, lateral options generated
- Most valuable insight — what was not visible before this analysis
- Next move — one concrete action this unlocks
What Claude Does / What You Do
| Claude | You |
|---|
| Selects the right instrument based on the problem type | Provide the stuck problem or plan |
| Applies techniques to your specific situation (not abstractly) | Confirm whether the new frame resonates |
| Names blind spots and new perspectives explicitly | Decide which insight changes how you act |
| Generates lateral options as starting points for evaluation | Evaluate and build on those options |
Related Skills
second-order-thinker — for consequence chains once a new frame is chosen
assumption-extractor — for systematic assumption auditing
tradeoff-articulator — for mapping costs when multiple options emerge