# Your site, as a machine reads it

> A free tool on oomira.com. Type any web address and see what an AI agent actually receives when it opens the page.

A page can look finished to a person and be close to blank to an agent: rendered by script, no text in the HTML, no structured data, behind a login. This tool reads an address the way an agent fetches it and shows what came back.

## What it reports

- **The pages.** What a reader receives from the address, still shaped like a page: headings as headings, tables as tables, images as the text they offered, and gaps where the reader got nothing. The findings sit beside the block that caused them.
- **The machine files.** What the site publishes about itself at fixed addresses, such as robots.txt, llms.txt, its sitemap and the well-known files an agent tries on arrival. Each is shown as served, refused or absent.
- **Elsewhere.** The places that describe the company which it does not own, and whether an agent can reach them.

Every row answers the same question, asked three times: can a machine reach this.

## What it does not do

- It gives no score. The page either hands a reader words or it does not, and that is a thing to look at rather than a number to argue with.
- It does not run JavaScript or drive a browser, so it reports what a plain fetch receives and says so when a page looks like it needs rendering.
- It reads one address at a time and has no account behind it, so what an AI agent opened or cited about that company is not shown here. Those come from the questions Oomira asks for an account.

## Where this sits

The tool answers what a machine can read. Oomira itself gets you set up to edit your own site with Claude Code, by a live session or the remote guide, with changes on staging first and live when you say go. After the setup it watches the site and hands you ideas for what to ask Claude next: from a site check like this one over every page, from what AI agents tell a buyer who asks about you, and from your own ideas.

- The tool: https://oomira.com/machine
- The company, described for an agent: https://oomira.com/llms.txt
- Start your setup: https://oomira.com/book

Canonical HTML: https://oomira.com/machine