The Stephen Hawking Claude Skill

Install Stephen Hawking's way of thinking into your AI. His mental models, heuristics and signature questions, engineered as a native Claude Skill and a plain .md for any LLM.

The rest of this site explains what Stephen Hawking thought. This page is about how he thought, and about a way to put that thinking to work: a downloadable Claude Skill that loads his reasoning patterns straight into your AI.

The framework is called Radical Inquiry Despite Constraints, and it captures Hawking's distinctive method, not as quotes or trivia, but as a structured toolkit of mental models, decision rules and the kinds of question he habitually asked. Drop it into Claude, ChatGPT, Gemini or any model, and your AI starts approaching problems the way he did, rather than in its default voice.

The core idea

Hawking's working life was a long argument that limitation, handled well, sharpens thinking rather than blunting it. Unable to fill blackboards with equations, he learned to reason in pictures and thought experiments, to strip problems down to their essentials, and to chase the biggest questions precisely where existing theories broke down. The framework distils that into something you can actually apply.

The five mental models

The skill is built around five named models drawn from his documented approach:

The Wheelchair Advantage: turning a constraint into a different, sometimes better, way of working.
The Thought Experiment Telescope: using imagined scenarios to see where real instruments cannot reach.
The Simplification Imperative: treating the clear, simple version of an idea as the goal, not the starting point.
The Collaboration Multiplier: pairing complementary minds to go further than any could alone.
The Public Translation Test: judging whether you truly understand something by whether you can explain it to anyone.

Each comes with concrete examples and the decision heuristics behind it in the full framework.

A taste

One of the signature questions the skill teaches your AI to ask is: what happens at the boundary where our current theories break down? It is a small example of the framework's flavour, pushing toward the edge of knowledge rather than away from it.

The framework is defined as much by what it refuses as by what it does. Among the anti-patterns it rejects: abandoning a fundamental question simply because the tools to answer it do not yet exist, exactly the kind of caution Hawking spent his career ignoring.

What you get

The full skill is roughly two thousand words and includes the core philosophy, all five mental models with worked examples, the decision heuristics Hawking used, his signature questions, the complete list of anti-patterns, a copy-paste activation prompt, and a worked example applying the framework to a realistic problem. It is delivered in two formats: a native Claude Skill (a .zip you upload in Claude's settings or unzip into Claude Code) and a plain .md file that works in ChatGPT, Gemini, Claude Projects, or any agent system prompt. It arrives by email within about a minute of checkout, and costs $4.99, with bundles available if you want to build a wider thinking library.

How this fits with the free site

Everything on this website is free, including the editorial account of how Hawking's mind worked. The Claude Skill is the engineered, installable version of that idea: the same temperament, packaged so your AI can actually reason with it. If you have enjoyed the free pages and want to put Hawking's way of thinking to practical use, the skill is the natural next step.