Mental Models Decision Advisor

"The quality of our thinking is largely influenced by our mental models." — Shane Parrish

Workflow: How to Use This Skill

When this skill is triggered, follow these five steps. Do NOT just dump model descriptions — actively guide the user through their specific problem.

Step 1: ASK — Understand the Decision

Before selecting any models, ask the user (if not already clear):

• What specific decision or problem are you facing?
• What are the options you're considering?
• What constraints or context matter (timeline, resources, stakes, reversibility)?
• What have you already tried or considered?

If the user's message already contains enough context, proceed directly to Step 2.

Step 2: SELECT — Pick 2-3 Relevant Models

First, run semantic search with the user's problem as the query to find the most relevant models. Then cross-reference with the routing table below to ensure coverage.

npx tsx "$MM/search-models.ts" "<user's problem description>" --top 5 --json

Read the full markdown file for each top result before proceeding. Use the routing table as a secondary guide:

State which models you selected and why they fit this situation.

Step 3: APPLY — Walk Through Each Model

For each selected model, apply it directly to the user's situation:

• Name the model and its core principle (one sentence)
• Show what it reveals about the user's specific problem
• State the concrete insight or implication

Step 4: SYNTHESIZE — Combine Into a Recommendation

Merge the insights from all applied models into:

• A clear recommendation or ranked options
• Key factors that tipped the balance
• Conditions under which the recommendation changes

Step 5: CHALLENGE — Stress-Test With an Opposing Model

Pick one model that argues against the recommendation. Apply it honestly:

• What does this counter-model reveal?
• Does the recommendation survive the challenge, or does it need adjustment?
• State final confidence level and any caveats

Scripts

Before calling any script, resolve the scripts directory (version may vary):

MM="${CLAUDE_PLUGIN_ROOT}/skills/aio-mental-models/scripts"

$MM/list-models.sh                    # List all 54 models grouped by volume
$MM/list-models.sh --volume 1         # Filter by volume (1-4)
$MM/list-models.sh --search "thinking" # Search by keyword
$MM/list-models.sh --count            # Quick count

Semantic Search

Find relevant models by meaning, not just keywords. Uses pre-computed embeddings (snowflake-arctic-embed-xs, 384-dim, runs locally).

Search (dependencies auto-install on first run):

npx tsx "$MM/search-models.ts" "how to think about risk and uncertainty"
npx tsx "$MM/search-models.ts" "team dynamics and collaboration" --top 3
npx tsx "$MM/search-models.ts" "startup growth strategy" --json

Options:

• --top N — Number of results (default: 5)
• --json — Output as JSON for programmatic use

Important

Always run semantic search first before selecting models. The search uses embeddings to find the most relevant models for the user's specific problem — this is more reliable than guessing from the catalog or your training knowledge. After searching, read the full markdown file for each selected model.

Model Catalog

Volumes:

• Volume 1: General Thinking (First Principles, Inversion, etc.)
• Volume 2: Physics, Chemistry & Biology (Leverage, Catalysts, etc.)
• Volume 3: Systems & Mathematics (Feedback Loops, Compounding, etc.)
• Volume 4: Economics & Art (Incentives, Opportunity Cost, etc.)

Volume 1: General Thinking Concepts

Foundational models for clear reasoning and effective decision-making

Supporting Ideas:

• Falsifiability - Theories must be testable
• Necessity vs Sufficiency - Required vs Enough
• Causation vs Correlation - Don't confuse coincidence with cause

Volume 2: Physics, Chemistry & Biology

Models from natural sciences for understanding change, energy, and adaptation

Physics

Chemistry

Biology

Volume 3: Systems & Mathematics

Models for understanding complex systems and mathematical patterns

Systems

Mathematics

Volume 4: Economics & Art

Models for understanding value, markets, and human expression

Economics

Art

Quick Reference Card

The Top 10 Most Useful Models

1. First Principles - Break it down to fundamentals
2. Second-Order Thinking - Think ahead: "And then what?"
3. Inversion - Avoid failure instead of chasing success
4. Feedback Loops - Understand cause and effect cycles
5. Compounding - Small consistent actions win long-term
6. Margin of Safety - Always have a buffer
7. Opportunity Cost - Consider what you're giving up
8. Leverage - Find maximum impact points
9. Hanlon's Razor - Assume incompetence, not malice
10. Circle of Competence - Know your limits

Daily Checklist

Before any major decision, ask:

• Am I reasoning from first principles or by analogy?
• What are the second and third-order effects?
• What would I do if I wanted this to fail? (Inversion)
• What's the probability of success? What's my confidence level?
• What am I giving up by choosing this? (Opportunity Cost)
• Do I have adequate margin of safety?
• Is this within my circle of competence?

Sources & Further Reading

• Farnam Street - The Great Mental Models
• Farnam Street - Mental Models Guide
• James Clear - Mental Models
• ModelThinkers

"You only think you know, as a matter of fact. And most of your actions are based on incomplete knowledge and you really don't know what it is all about, or what the purpose of the world is, or know a great deal of other things. It is possible to live and not know." — Richard Feynman