The Paul Graham
Framework
Essays, Startups & Doing Things That Don't Scale
Paul Graham built capital over decades by applying specific, repeatable mental models to uncertainty. Paul Graham's investing thinking, captured in this framework as mental models, decision heuristics, and the signature questions they would ask.
The essays, startups & doing things that don't scale framework distils Paul Graham's documented approach into a .md skill file you can load into Claude, ChatGPT, or any LLM in seconds.
Where most AI personas generate generic output, this framework is grounded in real documented work: interviews, writing, published frameworks, and the specific decisions Paul Graham made when it mattered. Use it for decision-making, writing, strategy, or any situation where you want your AI to think with Paul Graham's distinctive patterns rather than a default model voice.
A glimpse of Paul Graham's thinking.
Most breakthrough ideas look obviously wrong to experts because they threaten existing systems. The path to scale runs through deliberate unscalable work—talking to users one-by-one, hand-curating experiences, doing things founders "shouldn't" do. Write to discover what you think, not to express what you already know.
- 01
- 02
- 03
- 04
- 05
Five specific models, each named and explained with concrete examples, unlocked with purchase.
What would I work on if I knew no one would ever pay attention to it?
4 more questions in Paul Graham's voice are unlocked with purchase.
Full list of anti-patterns unlocked with purchase.
The full framework adds: detailed explanations of all five mental models with real-world examples, the decision heuristics Paul Graham actually used, remaining signature questions, the complete list of anti-patterns, a copy-paste activation prompt, and a worked example.
Unlocked the moment you purchase. ~2,000 words total. Delivered as a Claude Skill and plain .md within 60 seconds of checkout.
- ✦Core philosophy: 2–3 sentences grounded in Paul Graham's documented worldview
- ✦Five signature mental models: each named, with concrete examples from their work
- ✦Decision heuristics: the concrete rules they actually used
- ✦Signature questions: five questions that sound like Paul Graham, not a generic MBA
- ✦What they would NOT do: the anti-patterns this framework rejects
- ✦Copy-paste activation prompt: drop straight into Claude or ChatGPT
- ✦Worked example: the framework applied to a realistic scenario
- →paul-graham.zip: a native Claude Skill. Upload to Settings → Skills in Claude, or unzip to
~/.claude/skills/ - →paul-graham-framework.md: plain markdown. Works in ChatGPT Custom GPTs, Gemini, Claude Projects, agent system prompts, anywhere
What it does with a real question.
A founder is deciding whether to take a large but difficult client.
Take the client, but only if you treat it as a classroom. The word "difficult" is doing a lot of work in your description, and difficulty is usually where the information lives. Most founders flinch from a demanding client because the work is tedious and doesn't look like a product. That flinch is what Paul Graham calls schlep blindness, and it's often the sign you're near something real. A big client who makes your life hard is showing you, for free, what every future customer in that segment will eventually demand.
So the question isn't whether the revenue is worth the pain. It's whether you'll learn things your competitors can't learn because they declined. If their problems are specific to them, a one-off mess, walk away. If they're the sharp version of problems a hundred similar companies have, this client is a gift.
Here the danger is real, though: a large client can quietly turn you into their consulting shop, bending every decision toward their wishes. Guard against that by doing the hard work by hand, personally, for the first few months. Don't build features for them yet. Sit in their workflow, answer their emails yourself, and write down what keeps recurring.
Next move: this week, agree to a ninety-day paid pilot with a hard scope, and block out time to spend two full days inside their operation. At the end, you'll know whether you have a customer or a captor.
The full framework carries five models, decision heuristics, signature questions and a worked example.
Written by claude-sonnet-5-5 with this framework loaded, 2026-10-11. Checked automatically for length and for copied framework text.
Save up to 40% with a bundle
Works as a Claude Skill (.zip) and as a plain .md for ChatGPT, Gemini & every LLM
Delivered to your email within 60 seconds · Secured by Stripe
Price includes UK VAT where applicable
Ten frameworks for allocating capital and attention
Get Paul Graham plus nine more for $29.99 →Inspired by the documented thinking of Paul Graham. A mental-model toolkit, not a literal representation. Delivered as a .md file compatible with Claude, ChatGPT, Gemini, and any LLM.