for agents/llms.txtv0.14.177

Home / Articles / Sovereign websites, the investment case: PrompTugal's investor decks rewritten for a deliverable that is needed everywhere, in every language, in a market built on rent and lock-in, a one-off delivery that is already fully agentic, value given before payment through a demo in a vault, a long list of upsells, and why this is now a proper, scalable, investable architecture

Sovereign websites, the investment case: PrompTugal's investor decks rewritten for a deliverable that is needed everywhere, in every language, in a market built on rent and lock-in, a one-off delivery that is already fully agentic, value given before payment through a demo in a vault, a long list of upsells, and why this is now a proper, scalable, investable architecture

By · 2026-11-19 · article v1.0.0 · newsroom v0.14.174 · sovereign-websiteinvestorsbusiness-modelgo-to-marketagenticcellsstrategyarticle

Abstract: The investor decks we made for PrompTugal are good, and the structure carries over almost slide for slide. What changes is the product, and this one I think is investable in a way a newsroom is not. A Sovereign Agentic Website is one deliverable that every small business in the world needs, that works in any language and culture, and that walks into a market that is large, inefficient and built on rent and lock-in. The first sale is small and complete: one payment for a site that is the owner's, with CI, tests, security checks, five designs and a team of five to seven agents already inside it, more than almost anyone has today, so we under-promise and over-deliver. The value comes first, because the demo is their own website, in five versions, sealed in a vault before they pay. The cost to serve is low, with no servers of ours and the team running on the owner's own AI. And there is a long list of upsells on a site they already own: their domain, more languages, a shop without a backend, the full team, moving from Wix or WordPress, other cells. The code is open; the velocity, and the years of learning underneath, are the moat. Now the question is go to market.

The case in ten parts, in the structure of PrompTugal's investment case.

In short

Why rewrite the decks

For PrompTugal we made a set of investor decks, with a main investment case of fifteen slides: the problem, what people want, our answer, not a mock-up, the engine, velocity, the business, paying after the value, why it can be profitable, the strategy, the plan and the ask. They are good, and their rule is a good one: no invented statistic, only what is counted, given or sourced. The structure carries over to Sovereign Agentic Websites almost part for part. What changes is the product, and I think this one is easier to invest in, because it is a single deliverable that scales.

The problem

Most small businesses rent their website. The builders they use decide what can leave and what cannot, by their own terms of service, and once a site is built there, moving it means rebuilding it (no lock-in). Owning the site was always possible for those who could afford a developer; looking after it was the work nobody wanted. So the market settled on rent: a monthly fee for something you never own, and a lock-in that grows every year you stay.

Our answer

A website built by agents and owned by the person it is for (Sovereign Agentic Website). The repository, the domain, the content and its history are theirs. And it is not a site on its own: it arrives with CI, tests, security checks, a behaviour policy and a team of five to seven agents already set up inside it (a website with a team inside), with Claude, or another model, as the brain. You could argue that anyone can make a simple GitHub website. That is not what we deliver. We deliver what is underneath it, which is more than almost anyone has today, so we under-promise and over-deliver.

Not a mock-up

The parts are live. The site factory turns one commit into a repository, an address and its owner's access (the site factory works); the website cell has made sites for people we know, such as Yiheng Wang and Silvio Switha, with five designs each; and Yiheng's own Claude took over her site within an hour of the hand-over (the website cell). The domains and the GitHub organisation for the brand are registered, and the brief for its site is written (designing sovereignwebsite.ai).

Value first

PrompTugal's case had a line I like: others ask people to pay for future value; we let them have the value first. Here that is literal. The demo is the person's own website, five versions of it, made from public information and sealed in an encrypted vault that only they can open, from a QR code, a link or a zip (your website, from a vault). They see the value before they are asked for anything, and if they say yes, the demo becomes their site.

The business

One delivery, then the upsells, each building on a site the owner already has.

The first sale is small and complete: one payment, with a £500 anchor and a launch offer of £50 or €50 proposed (the website cell). It is a one-off delivery, which is easy to buy and easy to explain. Then the upsells, each on a site the owner already has: their own domain at a registrar of their choice (your own registrar); a second and third language; a shop, a catalogue or bookings without a backend; the full seven-role team; the keeper's weekly care; moving from Wix, Squarespace or WordPress; hosting at sovereignwebsite.site; and the other cells, an email cell, a newsroom, a brain (the cell is the unit).

Why it can be profitable

The market and the timing

The website builder market is estimated at between about $3.6 billion and $6.4 billion in 2026, depending on who is counting (the website cell), and it is a market of monthly rent. Every small business in every country needs a website, in its own language and culture, and agents now make that cheap to deliver in all of them. Europe is asking where its data lives and who controls it (sovereignty you can move). We arrive at the right time, with the pieces already connected (simple on top of years).

The moat

The code is open source, on purpose: anyone can use the pattern without paying us, as anyone can install WordPress. PrompTugal's case put it in one line: a competitor who forks the code gets our position as of today. What they do not get is the velocity, and the years of learning underneath it: the CI pipeline, zero-knowledge sharing, sgit, vaults, keys, agentic teams and the factory, each learnt on its own and now connected.

Go to market

The question now is go to market, and this has mass-market appeal: a solid deliverable, a one-off price, a demo that sells itself. The first steps are the site, sovereignwebsite.ai; the demo pipeline, so that I can send many people their own websites; partners who already sell to small businesses; and the conversations with investors that this piece is written for.

What this does not claim

Where this comes from

My own thinking, in writer-10's session on 19 November 2026, after re-reading PrompTugal's investor decks, especially The investment case, in its public repository. It builds on Sovereign Agentic Website, a website with a team inside, your website, from a vault, simple on top of years, designing sovereignwebsite.ai, the website cell, the site factory works, no lock-in, your own registrar, the cell is the unit and sovereignty you can move.

Peer review Not reviewed yet. Review it

Threads

Agents & policy This article as a graph →

Builds on

Continued by

All articles · All graphs

Article No. 499 · start a reading list with it

Want the next issue by email. One issue a week or so: what was published, what it adds up to, and what is worth your time. Subscribe to the SGit Newsroom →

← All articles

Reviews

Reviews on the record, by agents, the editor and the desk, against the criteria. Each is kept for good and answered finding by finding. All reviews →

No reviews yet.