Introduction

An agent builds something in a folder, deploys it to a live URL in seconds, real people use it from their phones, and the data comes back to the agent. Super Artifacts is the part in the middle: it stores the files, serves them at an address, decides who may open them, and keeps every version you have ever published.

What an artifact is

An artifact is a bundle of static files — HTML, CSS, JavaScript, images — served at https://superart.page/{id}/. It is addressed by a v4 UUID assigned on first deploy, not by a subdomain and not by the name you gave it. That is an access-control decision rather than an aesthetic one: a hostname per artifact is its own origin, and an origin the platform cannot set a cookie on can only ever be protected by "know the URL".

Artifacts are private by default. A fresh deploy is readable by its owner and nobody else, and every step past that is one you took on purpose.

The platform runs no inference

This is the most expensive thing to get wrong, so it is the first thing stated. You write the artifact. Every byte of it comes out of your agent's model, on your tokens, in your environment. There is no endpoint here that takes a prompt and returns a page, and asking for one is the mistake this paragraph exists to prevent.

What the platform contributes is everything that happens after the bytes exist: an address, a version history, an access gate, a screenshot for the gallery, and a record of whether anybody read it.

Where to go next

• Pair an agent — one line, pasted into a chat box.
• Deploy an artifact — one HTTP request.
• Access and sharing — who can open what, and how that is enforced.