Combine multiple mental models for richer analysis. Use for complex problems requiring multiple lenses, high-stakes decisions, or when single models leave blind spots.
Installs just this skill. Get the whole plugin for auto-invocation.
⚡ How it fires
How this skill gets triggered: by you, by Claude, or both.
Fires itselfClaude auto-loads it when your prompt matches the work.
You can call itInvoke it directly when you want it.
Slash command/thinking-model-combination
👁️ Context preview
The summary Claude sees to decide when to auto-load this skill.
Combine multiple mental models for richer analysis. Use for complex problems requiring multiple lenses, high-stakes decisions, or when single models leave blind spots.
📊 Stats
Stars863
Forks126
LanguageJavaScript
LicenseMIT
📦 Ships with thinking-skills
</> SKILL.md
thinking-model-combination.SKILL.md
---name: thinking-model-combination
description: Combine multiple mental models for richer analysis. Use for complex problems requiring multiple lenses, high-stakes decisions, or when single models leave blind spots.
---# Model Combination
## Overview
Real-world problems rarely fit neatly into a single mental model. Model combination uses multiple frameworks together—sequentially, in parallel, or nested—to achieve deeper understanding than any single model provides. The skill is knowing how to combine models productively without creating confusion or analysis paralysis.
**Core Principle:** Multiple lenses reveal what single lenses miss. But combination requires discipline, not just accumulation.
## Read This First: The Anti-Patterns
Most combination attempts fail by *adding* rather than *integrating*. Internalize these limits before doing anything else:
- **Hard cap: 3-4 models, each with a distinct, named role.** More than that is "Model Soup" — contradictory conclusions, analysis paralysis, no recommendation. If you can't say what unique question each model answers, drop it.
- **Add a model only to cover a specific, named blind spot** — not to look thorough. If a second model just confirms the first, it added nothing (Checkbox Combination).
- **Don't blend incompatible worldviews** (e.g., Effectuation's "embrace uncertainty" + detailed prediction). Use them in sequence for different phases, or as a deliberate adversarial pair — never mashed together.
- **Decide how the models relate *before* applying them**, and pick a tiebreaker model up front for when they conflict.
If a single model already answers the question, use it alone and stop. Combination is the exception, justified only by high stakes or genuine multi-domain spread. (Full anti-pattern detail is in the "Combination Anti-Patterns" section below.)
## When to Use
- Complex problems spanning multiple domains
- High-stakes decisions where blind spots are costly
- When single models leave important questions unanswered
- Validating conclusions through different frameworks
- Teaching comprehensive analysis
- Building robust decision processes
Decision flow:
```
Analyzing a problem?
→ Does one model fully address it? → yes → Use single model
→ Are there important blind spots? → yes → ADD COMPLEMENTARY MODEL
→ Are stakes high enough to justify deeper analysis? → yes → USE MULTIPLE MODELS
```
## Combination Patterns
### Pattern 1: Sequential (Pipeline)
Use one model's output as another's input:
```markdown
## Sequential Combination
Model A → Model B → Model C
Example: Product Decision
1. Jobs to be Done → Identify the real user need
2. First Principles → Design solution from fundamentals
3. Pre-mortem → Identify what could go wrong
4. Reversibility → Assess if we can course-correct
- [ ] Models are compatible or deliberately adversarial
- [ ] Synthesis addresses convergence and divergence
- [ ] Not using more models than necessary
- [ ] Clear combined conclusion emerges
## Key Questions
- "What does each model contribute that others don't?"
- "How do these models relate to each other?"
- "Where do the models agree? Disagree?"
- "Am I adding models for insight or just thoroughness?"
- "What's the simplest combination that addresses the problem?"
- "How do I synthesize if models conflict?"
## Munger's Wisdom (Extended)
"I've long believed that a certain system—which almost any intelligent person can learn—works way better than the systems most people use. What you need is a latticework of mental models in your head."
"You may have noticed students who just try to remember and pound back what is remembered. Well, they fail in school and in life. You've got to hang experience on a latticework of models in your head."
The latticework isn't just having models—it's the connections between them. Combination is how you weave the lattice. Individual models are threads; combination creates the fabric that catches reality's complexity.
