Quick Start • The /founder Command • All Commands • Multi-Platform
You know that thing where you're staring at a business problem at 2am and thinking "there's probably a framework for this"?
There are 568 of them. And now they live in your terminal.
Founder Skills gives you instant access to the best thinking from Porter, Christensen, Reid Hoffman, Brian Balfour, Teresa Torres, Paul Graham, and dozens more — as AI-powered slash commands that actually apply the frameworks to your specific situation. Not generic advice. Not Wikipedia summaries. Real analysis, with assumptions called out and recommendations you can test.
Works with Claude Code, OpenAI, Gemini, and any LLM.
"Like having McKinsey, YC, and a16z in your terminal, except they give you straight answers and don't charge $500/hr."
One command. That's it.
/plugins marketplace TasteDotDev/founder-skills
Install the founder-skills plugin when prompted. Done. You now have 16 slash commands and 568 frameworks.
Check what's installed anytime:
/plugins
Prefer git clone?
# Global install (all projects)
git clone https://github.com/TasteDotDev/founder-skills ~/.claude/skills/founder-skills
# Project-only install
git clone https://github.com/TasteDotDev/founder-skills .claude/skills/founder-skillsNow try it:
/founder We're a 50-person B2B SaaS doing $5M ARR, growth slowed from 3x to 1.5x
That's it. /founder figures out you need Four Fits + Growth Accounting + Unit Economics, applies all three to your situation, and gives you a prioritized action plan with validation steps. No framework selection needed.
/founder is the one command to rule them all. Describe any business problem in plain English and it will:
- Diagnose what you're actually dealing with (not what you think you're dealing with)
- Select the 2-4 best frameworks from 568 options
- Apply each one with full rigor — every section, every dimension, no shortcuts
- Challenge its own analysis (devil's advocate, assumption stress-test, pre-mortem)
- Synthesize into a prioritized action plan with specific validation gates
/founder Should we pivot from B2C to B2B?
/founder I need to pitch investors for our Series A
/founder Our team is growing from 10 to 50, how should we restructure?
/founder Competitor just raised $100M and is undercutting our prices
/founder How will AI agents disrupt our industry in the next 3 years?
The secret sauce: Every analysis ends with an assumptions table and a "here's how to test this before committing" section. Because the honest answer to most strategy questions is "it depends, and here's how to find out."
112 real prompts, real Founder CLI output. Here are a few:
founder "I have 200 beta users for my AI tutoring app. 30% come back weekly but nobody's paying yet. How do I know if I have PMF?"
founder "We're raising pre-seed for our AI code review tool. $10K MRR, growing 25% MoM. Need help with our pitch deck."
founder "Our competitor just raised $50M. We're bootstrapped at $400K ARR. What's my playbook?"
founder "We built a sales email generator on GPT-4. What's our moat against OpenAI and clones?"
founder "50K free users, 1.2% paid conversion, $12/mo price point. How do I double conversion?"
Browse all 112 examples with real output →
/founder handles everything. Go direct to a category only when you know exactly what you need.
| Command | # | Frameworks |
|---|---|---|
/founder |
568 | The brain. Picks the right frameworks for you. |
/founder-strategy |
48 | Porter, Blitzscaling, Wardley Maps, Cold Start, AI GTM |
/founder-marketing |
54 | PLG, Build in Public, TikTok Growth, Dark Social |
/founder-innovation |
50 | Superhuman PMF, Amazon PR/FAQ, RICE/ICE |
/founder-finance |
46 | Sequoia Deck, SAFE Notes, Growth Accounting |
/founder-organization |
40 | Async-First, Team Topologies, Kotter Change |
/founder-decision-making |
39 | Decision Matrix, Cynefin, Pre-Mortem, OODA |
/founder-communication |
38 | Pyramid Principle, SCQA, Elevator Pitch |
/founder-productivity |
34 | GTD, Deep Work, OKRs, Atomic Habits |
/founder-leadership |
33 | Founder Mode, Radical Candor, GROW Model |
/founder-sales |
30 | Cold Email, MEDDIC, Land & Expand |
/founder-design |
27 | Design Thinking, UX Research, Service Blueprint |
/founder-operations |
25 | Lean, Six Sigma, Theory of Constraints |
/founder-mental-models |
23 | First Principles, Inversion, Second-Order Thinking |
/founder-negotiation |
22 | BATNA, Harvard Method, Anchoring |
/founder-psychology |
14 | Nudge Theory, Cognitive Bias Audit, Flow State |
It's not a chatbot that read a business book once. Every framework has:
- A structured expert prompt — the AI adopts the mindset of a specialist in that exact domain
- Required inputs — it asks you the right questions, not generic ones
- Full output structure — every section, every dimension, no hand-waving
- Assumptions audit — every analysis surfaces what it assumed and how to validate it
- Contrarian check — the AI argues against its own recommendation to stress-test it
The result reads like something you'd get from a top-tier strategy consultant, except it's honest about what it doesn't know.
|
|
|
|
Plus classics: Porter's Five Forces, Jobs To Be Done, Design Thinking, OKRs, SWOT (but only when it's actually the right tool), and 477+ more.
Founder Skills works everywhere, not just Claude Code.
git clone https://github.com/TasteDotDev/founder-cli && cd founder-cli
python3 build.py| Platform | Instructions | Framework details |
|---|---|---|
| GPTs | dist/openai/gpts/<category>/instructions.md |
Upload knowledge/frameworks.md as knowledge file |
| Assistants | dist/openai/assistants/<category>.json |
Upload files/<category>-frameworks.md to vector store |
| Platform | Instructions | Framework details |
|---|---|---|
| Gems | dist/gemini/gems/<category>/instructions.md |
Upload knowledge/frameworks.md as knowledge file |
| API | dist/gemini/api/<category>.md (all-in-one) |
Included inline |
Individual self-contained prompts at dist/standalone/<category>/<framework>.md — copy-paste into any LLM.
python3 build.py # Build all platforms (640 files)
python3 build.py --clean # Clean rebuildOutput structure
dist/
openai/
gpts/<category>/instructions.md + knowledge/frameworks.md
assistants/<category>.json + files/<category>-frameworks.md
gemini/
gems/<category>/instructions.md + knowledge/frameworks.md
api/<category>.md
standalone/<category>/<framework-slug>.md
manifest.json
Each skill has two files:
skills/strategy/
├── SKILL.md # System prompt + framework catalog
└── frameworks.md # Detailed expert prompts for each framework
SKILL.md is what runs when you invoke a slash command. frameworks.md is the deep reference the AI reads when applying a specific framework — expert persona, required inputs, output structure, origin attribution.
These frameworks come from decades of the best business thinking:
| Domain | Thinkers |
|---|---|
| Strategy | Michael Porter, Clayton Christensen, Reid Hoffman, Andrew Chen |
| Growth | Brian Balfour (Reforge), Sean Ellis, Wes Bush, Elena Verna |
| VC / Startup | Y Combinator, Sequoia, a16z, First Round, OpenView |
| Product | Teresa Torres, Rahul Vohra, Marty Cagan, Jeff Bezos |
| Leadership | Paul Graham, Kim Scott, Patrick Lencioni, Brian Chesky |
| AI | Anthropic, OpenAI, a16z AI team, Kyle Poyar |
| Psychology | Daniel Kahneman, Richard Thaler, Robert Cialdini |
| Communication | Barbara Minto, Chip & Dan Heath, Jonah Berger |
We built the prompts. They built the ideas. Credit where it's due.
More memes
Want to add a framework? PRs welcome.
- Add a JSON definition to
server/tools/definitions/<category>/<slug>.json - Run the skill regeneration script
- Submit a PR
MIT License • Copyright (c) taste.dev
Now stop reading READMEs and go build something.