---
title: Mental Models OS
category: product
entity_type: skill
price: $15
canonical: https://forgehouse.ai/skills/mental-models-os/
lang: en
hreflang_alt: https://forgehouse.ai/tr/skiller/mental-models-os/
last_updated: 2026-06-20
---

# Mental Models OS

> Orchestrate decision-making with 80 mental models organized as a lattice model selection…

A decision-making operating system built on 80 mental models organized as a lattice. It classifies your problem, selects the right two-to-four models, applies them in a sequenced chain, and enforces anti-overstacking discipline so you frame decisions clearly without falling into analysis paralysis.

## Use cases
- Framing a complex strategy or pivot decision with multiple variables
- Running root-cause analysis on a bug, regression, or churn spike
- Building hypotheses before an A/B test or audit
- Pricing and packaging decisions weighed across economic models
- Multi-domain ambiguity where one call touches several areas
- Security and risk decisions that demand premortem thinking

## Benefits
- Break single-model tunnel vision by cross-framing every decision
- Avoid analysis paralysis with a hard four-model ceiling per chain
- Make decisions transparent: each model carries a one-line justification
- Reach better calls faster with pre-built chains for common decision types

## What’s included
- An 80-model catalog across eight categories with definitions and use cases
- A top-ten deep dive of the highest-frequency models with worked examples
- A chain decision tree mapping decision type to a ready model sequence
- Eight pre-built chain templates for strategy, debug, pricing, churn, security, and more
- Anti-overstacking discipline with category-diversity and time-box rules
- A deterministic output schema capturing models, order, justification, and recommendation

## Who it’s for
Founders, strategists, and operators who make high-stakes, multi-variable decisions and want a disciplined framework instead of gut feel.

## How it runs
Eighty mental models help nobody until the right four are chained in order. Decisions get classified by type, matched to a chain template, and resolved with a visible reasoning trail instead of an opaque verdict.
1. Classifies the decision into one of eight types (strategy or pivot, debug or root cause, pricing, churn risk, security, content and marketing, system scaling, multi-domain ambiguity), because the type determines which models earn a seat.
2. Picks the matching chain template from the library: for example debug runs Hanlon's Razor then Occam then 5 Why then a Bayesian posterior, pricing runs Opportunity Cost then Marginal Utility then Anchoring then Comparative Advantage. If no template fits, it builds a custom chain capped at four models.
3. Runs the anti-overstacking gate before applying anything: maximum four models per chain, at least two different categories represented (an all-statistics chain is an echo chamber), and trivial problems are rejected outright because a typo fix does not deserve a premortem.
4. Applies the models sequentially, each model's output feeding the next as input: an Inversion premortem produces failure scenarios, the Bayes step anchors them against base rates, the Second-Order step traces what the fix itself will cause.
5. Aggregates the chain into one concrete recommendation and writes a one-sentence justification per model used, so the decision never arrives as an opaque verdict but as a visible reasoning trail.
6. Persists the chain history and the final decision to memory, building a searchable record of which model combinations were used on which problem class and what they concluded.

## FAQ
### Is this only for business strategy calls, or does it help with technical decisions too?
The pre-built chains cover both: root-cause analysis on bugs and regressions, security premortems, and A/B test hypothesis building sit next to strategy, pricing, and churn chains. The decision tree routes by decision type, not by department.

### How does it decide which of the 80 models apply to my problem?
It classifies the problem first, then a chain decision tree maps the decision type to a sequenced set of two to four models, each carrying a one-line justification. Eight pre-built chain templates handle the most common cases without you assembling anything.

### Will it make the decision for me?
No. The output schema captures the models used, their order, the justification, and a recommendation, but the call stays yours. It frames the decision clearly; it does not replace judgment.

## Price
$15, one-time, no subscription. VAT included.

Related guide: [AI for small business](https://forgehouse.ai/guides/ai-for-small-business/)