The documents every startup should have — compiled into a knowledge base an AI can actually read.
Drop your pitch deck, market research, business plan, competitor notes and call transcripts into one folder → an AI turns them into a structured, front-mattered knowledge base → then a library of prompts runs on top of it to produce the work: the deck, the list of 100, the unit economics, the weekly recap.
Every company, customer, metric and transcript used as an example in this repo is invented. Nothing here is drawn from any real client engagement.
New to working with AI · What the words mean · The infographics · The method · Build the knowledge base · Prompt library · For coaches · For programmes · Quickstart
10 knowledge-base sections · 19 prompts · 10 worksheets · 30 infographics · 18 pages of method · a portfolio layer for programmes · a published site, generated from these files
Most early founders are carrying their company in their head and in twelve unrelated files. The pitch deck says one number, the projections say another, the market research lives in a PDF nobody has opened since March, and every AI conversation starts from zero — so every answer is generic, and the founder pays for the model to re-read everything, every time.
Meanwhile the advice they need is not exotic. It is the same twenty things, asked in the same order, by every coach who has ever sat across from an early-stage founder: Who exactly is the customer? What does a unit cost you? Who else is doing this? How many prospects are on your list — five, or a hundred? What happens when the money runs out?
startup-stack makes those questions answerable from one place.
There is a second reader, and the repo now serves them too: the programme running a portfolio of these companies — an incubator, accelerator, studio or university venture programme, whose coaching team meets the same founders in recurring sessions and whose institutional memory currently lives in scattered notes. The page written for them is the deeper end of the same method. A founder can ignore it entirely.
┌──────────────────────────────────────────┐
│ │
_inbox/ │ stack/ │ work
─────── │ ────── │ ────
pitch deck │ INDEX.md ← the router │ pitch deck v2
business plan ──┼─▶ CONTEXT.md ← the one-pager ├─▶ list of 100
market research │ 01-company … 10-pulse │ unit economics
competitor notes │ ↑ one .md per section, front-mattered │ investor one-pager
call transcripts │ ↑ confirmed / unverified / TBD / clash │ weekly recap
spreadsheets │ │
design refs │ │
└──────────────────────────────────────────┘
LOOP 1 · BUILD LOOP 2 · ENRICH LOOP 3 · PULSE
(one afternoon) (every time something (30 minutes,
real happens) every week)
Loop 1 — Build. You put raw material in _inbox/. One prompt reads it and writes the stack: ten numbered sections, one markdown file each, every fact tagged confirmed, unverified, missing, or in conflict where two of your own documents disagree. Takes an afternoon. The output is deliberately incomplete — the gaps are the point, because a gap you can see is a task.
Loop 2 — Enrich. Every real thing that happens — a customer call, a supplier quote, a rejected ad campaign, a new competitor — goes back into the stack. The base is never finished. Each addition makes the next request cheaper and better, because the AI stops guessing about your business.
Loop 3 — Pulse. Once a week, one prompt reads the stack, reads what changed, and produces a recap: what moved, what did not, what you learned, what you are doing next, and how many weeks of money you have left. That recap is the file you send your coach, your co-founder, or your investors — and it is also the file that gets read back into the stack next week.
| Folder | What it is |
|---|---|
stack/ |
The templates. Ten numbered sections — company, customer, market, product, GTM, operations, money, capital, brand, pulse — one markdown file each, with front matter. stack/INDEX.md is the router an AI reads first, and stack/CONTEXT.md is the one-pager it reads next. This becomes your knowledge base; you fill it in, you do not copy it out. |
prompts/ |
The prompt library. 19 prompts: bootstrap, gap scan, competitive intelligence, unit economics, the list of 100, the pitch deck, fundraise readiness, the weekly recap, meeting-to-actions, and the adversarial review that tells you what an investor will attack. Each opens with a Requires: line and stops if what it needs is missing. Most want a filled-in stack; 00, 14 and 15 run against an empty one, which is how a stack starts to fill. The last three belong to portfolio/ and never touch a founder's stack. |
worksheets/ |
The repeatable artifacts. Fill-in files you produce over and over: startup-prep, customer interview, competitor profile, two-week experiment, SOP entry, trade/order sheet, advisor scope letter, one-pager, list of 100, weekly recap. |
docs/ |
The method, in eighteen pages. New to any of this: AI basics and what the words mean — the terminal, an editor, Git, GitHub, connectors, model tiers — each in plain language with a diagram. The method: why it works, building the knowledge base, the front-matter schema, getting your material in source by source, and website to inbox. Running it: the weekly recap, for coaches, the safety rules for giving an AI access to your files, and on a schedule — recipes for Microsoft 365, Google Workspace, Zoom, Slack, and none of them. Two are for programmes only: for programmes and the exchange folder. And the infographics — thirty pictures across four pages, covering what AI is, the tools, the method and running a portfolio; take them for a workshop. |
portfolio/ |
Optional — for a programme, not a founder. The templates an incubator, accelerator or studio uses to run this across many companies: a router, a 1-to-5 maturity rubric per function, the cross-company theme scan, and a company folder to copy. Delete it if you are a founder. See docs/for-portfolios.md. |
_inbox/ |
Where raw material lands before it is processed. Git-ignored by default. |
tools/ |
Optional. scrape-site.mjs turns a website into citable markdown in _inbox/ — you never need it, it saves an hour of copy-and-paste (docs/scraping.md). make-og.mjs redraws the social card at the top of this page; run it if you fork and rename. A third tool lives in its own repository: site2deck samples a company's colours, fonts and logo off its website and builds a branded, offline, single-file slide deck — the object to put in front of a room once prompt 10 has written what goes in it. Same habits as this repo: zero dependencies, Node only, unknowns marked loudly rather than filled in, and a --public build that strips anything tagged internal and refuses to ship if one survives. |
web/ |
This project's own site, generated from the markdown in this repo. Also holds web/infographics/ — the thirty images the gallery shows — the ten hand-built SVG diagrams in web/assets/diagrams/, and the social card. Not part of your stack; delete it when you fork, but take the infographics first if you want them. |
Requirements: a paid AI subscription and a tool that can see your file system — the Claude desktop app, Claude Code, ChatGPT's Codex app, or any equivalent. Free chat tiers cannot do this work; they cannot read your folder.
If that sentence already lost you, read docs/ai-basics.md first, and keep docs/what-things-are.md open beside it for anything it assumes you know. Between them they explain what "a tool that can see your file system" means, what the terminal is, what Git and GitHub are, what the AI genuinely cannot see, and the one rule that never bends — with no code.
# 1. Take a copy. This is a template, not a dependency — you own what you make.
git clone https://github.com/fritzhand/startup-stack my-company-stack
cd my-company-stack && rm -rf .git && git init
# 2. Fill the inbox. Anything that describes the business.
# Pitch deck, business plan, market research, competitor notes, financial model,
# call transcripts, WhatsApp exports, supplier quotes, the website.
cp ~/Downloads/*.pdf _inbox/
# 3. Point your AI at the folder and run the bootstrap prompt.
# prompts/00-bootstrap-the-stack.mdThen read QUICKSTART.md — it walks the first hour honestly, including what not to try to finish today.
This is what it costs, measured on a real base rather than a demo:
Skeleton in an hour. A working v1 in a few focused sessions. A stack you actually trust in about a month, because the long pole is you correcting what the AI got wrong — not the AI writing it. Then roughly thirty minutes a week to keep it alive.
Nobody should tell you this is free. The AI drafts; you validate. A stack you have not corrected is a confident-sounding guess about your own company, and that is worse than no stack at all.
- The document is the data. If a metric needs special effort to gather, it is the wrong metric. Your numbers should fall out of records you already keep.
- AI drafts, the founder validates. Nothing in the stack is true because the model wrote it. Every section carries a status tag until a human has confirmed it.
- Tag honestly.
needs-verificationandtbdare not failures — they are the quality system. A stack with visible gaps beats a polished one that quietly invents numbers. - Enrich, don't perfect. Ship the skeleton. Add to it forever.
- One index, one summary line per file. This is what keeps the base cheap: the AI reads a small router and opens one section, instead of re-reading everything you own on every question.
- Private master, shared derived. Every file declares a sensitivity. What you share with a coach, an investor or a customer is carved from the master — never the master itself.
- Scope the AI to one folder. Start narrow. Widen only as fast as your confidence. See the safety rules.
The method is published at fritzhand.github.io/startup-stack — every page in docs/, every prompt, every worksheet, and this file, with search, a filterable prompt library, and light and dark themes.
The sidebar groups into seven sections, each with its own overview page listing what is in it. Diagrams drawn as SVG re-skin with the theme; the thirty infographics open full size in a viewer rather than navigating away from the page explaining them.
Generated from these files, not written twice. Each page is its markdown rendered; the home page is assembled out of this README's own sections. Edit the markdown and the page changes — and if a heading the home page depends on is renamed, the build fails rather than shipping an empty section.
What is deliberately not published: the section templates in stack/ and the company templates in portfolio/. Those are files you fill in with your own business, so the site lists them with their status instead of their contents. That matters if you fork this and publish your own — web/build.mjs withholds the summary: of any section that is not public once it has been filled in, and prints what it withheld.
node web/build.mjs # markdown + web/assets → _site/ (zero dependencies, Node 18+).github/workflows/pages.yml runs that on every push to every branch and on every pull request, so a broken page is caught before it is merged; only main publishes. The build fails loudly on a broken internal link, a missing diagram, a prompt not routed to a group or a page with no title, rather than shipping a quietly wrong page.
Site-wide settings live in site.config.json — name, tagline, base URL, repository, and analyticsId for a GA4 measurement ID. Leave analyticsId empty and no analytics tag is emitted at all; if you fork this and keep web/, change it or clear it, or your visitors are counted in someone else's property.
When you fork this as your own company's stack, delete web/ and .github/ — and portfolio/ unless you are a programme. They belong to this project, not to yours.
The knowledge-base method is not theoretical. It is adapted from the working practice of an incubator that runs its own operations — decks, reports, SOPs and monthly metrics — out of one structured folder of plain markdown files, rather than out of software.
The founder-advice content is standard early-stage curriculum: the questions an experienced coach asks, roughly in the order they get asked, and the answers that hold up. No client work, session material or client data is reproduced here. Every company, customer and number used as an example is invented for illustration.
A companion project, site2deck, does the last mile — it samples a company's real design system off its website and turns a filled-in stack into a branded, offline, single-file slide deck. stack/09-brand/brand.md is written to hand off to it cleanly.
MIT. Take it, fork it, rename it, run it with your cohort, sell coaching on top of it. Attribution appreciated, not required.