Choose the right mental model for the problem at hand. Use when facing new problems, when current approaches fail, or when you need to match tool to context.
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-selection
👁️ Context preview
The summary Claude sees to decide when to auto-load this skill.
Choose the right mental model for the problem at hand. Use when facing new problems, when current approaches fail, or when you need to match tool to context.
📊 Stats
Stars863
Forks126
LanguageJavaScript
LicenseMIT
📦 Ships with thinking-skills
</> SKILL.md
thinking-model-selection.SKILL.md
---name: thinking-model-selection
description: Choose the right mental model for the problem at hand. Use when facing new problems, when current approaches fail, or when you need to match tool to context.
---# Model Selection
> **Overlaps thinking-model-router.** If you already know which model fits, skip routing entirely and just invoke that model. If you don't, **thinking-model-router** is the single entry point — start there. This skill goes deeper on *how* to classify a problem and *when to abandon* a model once chosen. (Audit recommends merging this into thinking-model-router; for now, prefer the router as the front door.)
## Overview
Every mental model has a domain where it excels and domains where it fails. Model selection is the meta-skill of recognizing which model fits which problem. The expert doesn't just know many models—they know when to apply each one. Using the wrong model is often worse than using no model at all.
**Core Principle:** The map is not the territory. Choose the map that best serves your journey.
## When to Use
- Facing a new problem type
- Current approach isn't working
- Multiple stakeholders suggest different frameworks
- Deciding how to structure analysis
- Teaching others which tools to use
- Building decision-making processes
Decision flow:
```
Facing a problem?
→ Have you identified the problem type? → no → CLASSIFY THE PROBLEM FIRST
→ Does your usual model fit? → no → CONSIDER ALTERNATIVES
→ Are you using a model by habit? → yes → QUESTION THE FIT
```
## When NOT to Use
- **You don't know which thinking tool to reach for.** Start with `thinking-model-router` — it is the single entry point that routes to the right skill. This skill goes deeper on *how* to classify problems and match models, but only after the router has narrowed the field.
- **You've already classified the problem type.** If you know it's a debugging problem, just invoke `thinking-scientific-method` or `thinking-five-whys-plus` directly. Don't run a full model-selection exercise to confirm what you already know.
- **The problem is routine and your default approach fits.** Model selection adds overhead; on a familiar problem, use the model you know works.
- **You're tempted to run this as a first step on every task.** That turns a meta-skill into a tax. Most tasks don't need model selection — they need execution. Reserve this for when your usual approach has demonstrably failed.
> **Redirect:** For nearly all cases, start with `thinking-model-router` — it classifies the problem and dispatches to the right skill in one pass. This skill is the reference for *how* that classification works, not a replacement for the router.
## Trigger Card
When your usual problem-solving approach has demonstrably failed and you need to match a different tool to the task:
- Parallel: Use A and B independently, compare results
- Nested: Use A at macro level, B at micro level
```
### When to Abandon a Model
```markdown
## Model Exit Criteria
Stop using a model when:
- 15+ minutes with no insight
- Key facts don't fit the framework
- You're forcing the analysis
- Someone suggests a better fit
Don't stop just because:
- The answer is uncomfortable
- It requires more work
- Results disagree with intuition (investigate why)
```
## Verification Checklist
- [ ] Classified the problem type
- [ ] Considered multiple candidate models
- [ ] Checked for model-problem fit
- [ ] Avoided using model out of habit
- [ ] Have a fallback if model doesn't work
- [ ] Can articulate why this model fits
## Key Questions
- "What type of problem is this really?"
- "What model would an expert in this domain use?"
- "Am I using this model because it fits, or because I know it?"
- "What aspects of the problem does this model ignore?"
- "What model would give a different answer?"
- "When should I switch to a different model?"
## Munger's Wisdom
"You've got to have models in your head. And you've got to array your experience—both vicarious and direct—on this latticework of models."
"You must know the big ideas in the big disciplines, and use them routinely—all of them, not just a few."
The power isn't in any single model—it's in having many models and knowing when each applies. A tool is only useful when matched to the task. The meta-skill of model selection multiplies the value of every model you know.
