# Mental Models > Apply Charlie Munger's latticework of mental models to any problem. Use when user requests decision analysis, says "help me think", "apply mental model", mentions model names (inversion, bottlenecks, second-order thinking), or needs structured thinking frameworks. Use when this capability is needed. - Skill: `tomevault-io/mental-models` (Agent Skill, multi-file: 2 files) - Install (CLI): `npx skillmds@latest add tomevault-io/mental-models` - Raw SKILL.md: https://api.skillmd.com/api/skills/tomevault-io/mental-models/raw - Safety review: pending (external: skill-scanner PASS, skillspector CAUTION) - Works with: Claude Code, Claude.ai, OpenAI Codex - Category: AI & ML - Author: tomevault-io (https://skillmd.com/u/tomevault-io) - Updated: 2026-09-17 - Page: https://skillmd.com/skills/tomevault-io/mental-models --- # Mental Models Apply 98 cognitive frameworks from multiple disciplines to analyze problems, make decisions, and think more clearly. This skill is backed by the **`mental-models` CLI** — a single command that does model selection, lookup, and structured application. The CLI is the fast path; the fallback is reading files directly. Both work. Prefer the CLI. ## When to Activate - User names a specific model ("apply inversion", "use bottlenecks") - User asks "help me think through X" or "what model fits X" - User requests decision analysis, trade-off evaluation, or structured reasoning - User describes a complex/ambiguous problem and wants a framework ## Preflight: is the CLI available? Run once per session: ```bash mental-models doctor --json ``` **If it returns `{"ok": true, ...}`** → use the CLI workflow below. **If the command is not found** → try `uvx mental-models doctor --json` (runs from PyPI without install). If that also fails, fall back to the **File Fallback** section at the bottom of this doc — you can still do everything by reading files directly from `models/`, `REFERENCE.md`, and `PATTERNS.md`. ## CLI Workflow (preferred) ### Step 1 — Select models for the problem ```bash mental-models select "" -k 5 --json ``` Returns a JSON object with a `models` array. Each entry has `slug`, `name`, `category`, `description`, `keywords`, `path`. Pick **2–3** that best fit — prefer cross-category coverage (that's the latticework). ### Step 2 — Get structured guidance for each chosen model ```bash mental-models apply --problem "" --json ``` Returns: - `description` — what the model is - `thinking_steps` — the sequential framework (walk these verbatim, don't paraphrase) - `coaching_questions` — prompts to deepen the analysis - `when_to_avoid` — failure modes (always check and surface if relevant) ### Step 3 — Synthesize - Walk each model's `thinking_steps` against the user's facts - Show where the models agree, where they disagree - End with 3–5 concrete, actionable next steps - Name any "when to avoid" conditions that apply to this case ### Other useful CLI commands ```bash mental-models get # full markdown for deep reading mental-models get --field keywords mental-models list --category "Human Nature" mental-models categories mental-models which # resolve data path ``` All commands support `--json`. Exit codes: 0 ok, 2 not found, 3 bad args. ## Discovery Heuristics (before calling select) Match the problem's **shape** to bias your query terms: - **Risk / uncertainty / reversibility** → inversion, probabilistic thinking, margin of safety - **Stuck / can't see options** → first principles, second-order thinking, reframing - **Conflict / negotiation / competition** → incentives, asymmetric warfare, trade-offs - **Complex system / unintended effects** → feedback loops, emergence, bottlenecks, leverage - **Performance / optimization** → bottlenecks, diminishing returns, efficiency - **People / team / behavior** → incentives, social proof, biases - **Communication / persuasion** → framing, audience, contrast Full decision trees: **PATTERNS.md**. Per-category deep walkthroughs: **REFERENCE.md**. Worked examples: **examples/**. ## Core Guidelines 1. **Max 3 models per analysis** — quality over quantity 2. **Follow `thinking_steps` verbatim** — don't paraphrase the framework away 3. **Always check `when_to_avoid`** — warn the user if the model misfits 4. **Latticework**: show how chosen models connect and where they disagree 5. **Be actionable**: end with concrete next steps, not theory 6. **Name biases honestly**: if the user seems caught in one, surface it ## Category Map | Category | IDs | Focus | |---|---|---| | General Thinking | m01-m09 | Foundations: inversion, first principles, second-order | | Science | m10-m29 | Natural laws: leverage, inertia, activation energy | | Systems Thinking | m30-m40 | Constraints, feedback, emergence, scale | | Mathematics | m41-m47 | Randomness, regression to mean, sampling | | Economics | m48-m59 | Scarcity, trade-offs, supply/demand | | Art | m60-m70 | Framing, audience, contrast | | Strategy / Warfare | m71-m75 | Asymmetric advantage, seeing the front | | Human Nature | m76-m98 | Biases, incentives, social proof | ## Files in This Skill - `SKILL.md` — this entry point (CLI-driven playbook) - `REFERENCE.md` — deep per-category walkthrough (fallback + teaching) - `PATTERNS.md` — decision trees for common problem shapes - `examples/` — 5 worked scenarios - `models/` — 98 model files (the source of truth the CLI reads) - `resources/model-index.json` — searchable keyword index - `resources/quick-reference.md` — problem→model lookup tables ## File Fallback (when CLI is unavailable) If `mental-models` is not installed and `uvx mental-models` is not available: 1. **Discovery**: read `resources/model-index.json` and grep `resources/quick-reference.md` for keyword matches 2. **Selection**: use the Discovery Heuristics above + **PATTERNS.md** decision trees 3. **Application**: open the model file at `models/Mental_Model_/m_.md` and walk the **Thinking Steps** section verbatim 4. **Always check** the **When to Avoid** section before recommending the model This fallback gives you the same content as the CLI — the CLI just makes selection, lookup, and section extraction faster and more deterministic. --- > Converted and distributed by [TomeVault](https://tomevault.io/claim/cyperx84) — claim your Tome and manage your conversions.