thinking-model-router

Routes analytical problems to the best-fit thinking skill by domain and problem type.

Updated May 24, 2026
One-click install
npx skills add https://github.com/MWest2020/skill-forge --skill thinking-model-router-mwest2020
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: thinking-model-router
Source: https://github.com/MWest2020/skill-forge/tree/main/skills/thinking-model-router
Command: npx skills add https://github.com/MWest2020/skill-forge --skill thinking-model-router-mwest2020

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve? When facing an analytical task, it is often unclear which mental model or thinking skill applies, leading to guessing, habitual tool-stacking, or forced frameworks. This Skill maps the problem's domain and type to a single primary thinking skill (or NONE) so the right reasoning frame is applied without over-application. ## Core Features & Use Cases - Domain and type classification: Characterizes problems across domains (coding, architecture, product, strategy, personal, risk, innovation) and types (diagnose, decide, understand, create, evaluate, predict, optimize). - Weighted selection criteria: Scores candidates on problem fit, available inputs, time to apply, stakeholder usability, and competence, returning NONE when no route clears the threshold. - Controlled multi-skill routing: Allows at most three complementary skills, each with a distinct role and unique question, with a domain cheat-map for fast lookups. - Use Case: When deciding whether to rewrite a legacy service, invoke the router to characterize the problem as architecture/decide, and it routes to reversibility or opportunity-cost rather than stacking every framework you know. ## Quick Start Ask the assistant to use the thinking-model-router skill to pick the right thinking skill for your current problem, describing the domain and what you need to figure out.

Frequently Asked Questions about thinking-model-router

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I choose the right mental model for a problem?

Characterize the problem by domain (coding, strategy, product, etc.) and type (diagnose, decide, create, evaluate), then match against type defaults such as five-whys-plus for diagnosis or probabilistic for prediction. Score candidates on problem fit and constraints, and return NONE if nothing clearly fits.

How do I combine multiple thinking frameworks on one problem?

Combine at most three skills, and only when each has a distinct complementary role answering a unique question the primary leaves open. Prefer sequential application over parallel, and never stack near-neighbor or synonymous frameworks.

When should I not use a thinking skill router?

Skip routing when the matching skill is already obvious, when the task is routine implementation with no analytical unknown, or when you are mid-execution of an agreed plan. Invoke the known skill directly or reason without a framework.

What happens if no thinking framework fits my problem?

The router returns NONE, meaning you reason directly without invoking a skill. A route must score at least 4 on problem fit and 3.5 weighted overall; otherwise NONE is the correct output and no frame should be forced.

Why does the router limit selection to three skills maximum?

The cap prevents framework stacking where overlapping models add noise instead of insight. Each additional skill must answer a distinct open question; near-neighbors and synonyms do not stack, and multi-skill routes collapse to the single best primary if roles are not distinct.