Munger — Mental Model Lattice
Take an investment idea and run it through Charlie Munger's multidisciplinary latticework. The goal is to see what pops: which mental models fire for the idea, which fire against it, and what blind spots an inversion pass exposes. This is a thinking tool to pressure-test a thesis — not financial advice; never tell the user to buy or sell, surface the reasoning.
The full model catalog lives in
(bundled next to this file). Read it at the start of every run. It is intentionally a growing list — when the user adds or refines models, edit
, not this file.
Workflow
1. Get the idea
Take the investment idea the user provides — a pasted thesis, a ticker, or a business description. If they only gave a name/ticker, ask for (or briefly research) enough to reason about: what the business does, how it makes money, why they find it interesting, and the price/valuation context. Don't proceed on a one-word prompt.
2. Load the lattice
Read
for the current catalog of models (grouped by discipline). Treat it as the checklist to run the idea against — Munger's whole method is running every idea past a
checklist of models rather than reaching for one.
3. Run the idea through the models — find what pops
For each model that is genuinely relevant, record:
- Model — name + the discipline it comes from.
- Direction — does it fire for the thesis, against it, or reframe it?
- So what — the specific, concrete read on this idea (not a textbook definition). Quote the part of the thesis it bears on.
Rank by how strongly each model fires. Surface the dominant few rather than listing all — "what pops" means signal, not a full sweep dump.
4. Inversion pass
Run Munger's "invert, always invert":
how does this idea fail? What would have to be true for this to be a terrible investment? Name the disconfirming models and the bear case the thesis is glossing over. Separately, flag any
psychological misjudgment tendencies (see the psychology section of
) that may be biasing the user's own thesis — e.g. commitment/consistency, social proof, deprival-superreaction.
5. Apply Munger's core filters
Explicitly check the four-filter screen:
- Circle of competence — can this be understood well enough to judge?
- Durable competitive advantage — is there a real moat, and is it widening or eroding?
- Able and honest management — capital allocation, incentives, candour.
- Margin of safety / sensible price — is the price demanding optimism, or leaving room to be wrong?
6. Verdict
Synthesize: which models dominate the picture, the two or three things that most need to be true, the biggest disconfirming risk, and what to investigate next. End with a disposition framed as thinking — "compelling on moat + price but hinges on X", "pass — outside circle of competence", "needs work — thesis leans on social proof" — never a buy/sell directive.
Rules
- Not financial advice. Output reasoning and disposition, never a directive to trade.
- Specific, not generic. Every model that fires must be tied to a concrete fact in the idea — quote it. No textbook recitations.
- Surface signal. Rank and feature the models that pop hardest; don't dump the entire catalog.
- Always invert. The inversion + bias pass (step 4) is mandatory, not optional.
- The catalog in is the source of truth for models — keep it growing there.