Builds and ships a website automatically from public information
Sites that generate themselves from public data
The pipeline finds a business, builds it a real website from information that's already public, and puts a live preview link online — no designer, no developer, no kickoff call.
An agency building each site by hand → a pipeline that generates and ships them.
| Before | After | |
|---|---|---|
| Who builds the site | a designer and a developer | the pipeline, automatically |
| Getting started | a kickoff call and an intake process | nothing — it starts from public information |
| Time to something real | weeks of back-and-forth | a live preview link produced on its own |
| What the business sees first | a quote and a timeline | a finished version of its own site |
| Cost to get online | an agency bill many can't justify | close to nothing for the first draft |
Making a real website goes from a multi-week professional service to something the pipeline generates and ships on its own. A business can look at a finished version of its own site before anyone has been hired or paid — which collapses the cost and the friction of getting online to almost nothing. This is built, working software, run as an experiment in how cheap web presence can get.
What I built
A pipeline — built, working software that runs on its own — that produces a small business a real website without anyone hiring a designer or developer first. A "pipeline" here just means an automated chain of steps: each one hands its output to the next, start to finish, with no person in the middle.
- It finds the business. The pipeline identifies a business to build for and gathers what's already public about it — no form to fill out, no kickoff call.
- It writes and builds the site. From that public information it generates the copy and assembles the pages into a professional, on-brand site, with no hand-coding.
- It ships a live link. The result isn't a mockup — it's deployed as a working preview link the business can open and click through.
- It does this with no one in the loop. The whole sequence — discover, gather, generate, publish — runs automatically, so the first finished draft appears before any conversation has happened.
This is standalone software with its own deployment, not a workflow that runs inside Claude. It's a live experiment, so the open question is keeping the output consistently accurate and on-brand across very different businesses.
Why it matters
For a small business, the promise is simple: see a finished version of your own website before you've spent a dollar or sat through a sales call. That removes the two things that keep so many businesses offline — the cost and the friction of getting started.
The deeper shift is in the economics. Normally a real website is a multi-week paid engagement: an agency, a designer, a developer, weeks of coordination. Here it becomes a generated preview link produced automatically. When the delivery cost falls that far, getting a small business online stops being a budgeted project and becomes something that can simply happen — which is exactly what this experiment is built to test.
- 01Discover
The pipeline identifies a business to build for, working from information that's already public.
- 02Gather
It pulls together what's publicly known about the business — what it does, where it is, how it presents itself.
- 03Generate
It writes the copy and builds the pages into a professional, on-brand site, with no hand-coding.
- 04Deploy
It publishes the result as a live preview link — a working site, not a mockup.
- 05Show
The business can open the link and see a finished version of its own site before any conversation has happened.
The real unlock is collapsing the cost of getting online: when a business can see a finished version of its own site before a single conversation, what was a multi-week paid engagement becomes a generated preview link. It's an experiment in flipping the economics of small-business web presence entirely.