Can AI find and cite your site?
Agentify My Site is our free scanner. Give it a URL and it grades how findable and citable your site is by AI, and how legible it is to autonomous agents. You get a score out of 100, a breakdown of every check behind it, and a paste-ready prompt for each gap.
Your customers are asking an AI, not a search box.
Ranking well was a question of where you sat in a list of links. Being recommended is a different question entirely: whether the model reading your site can understand it well enough to say your name back to someone who asked.
Most sites were never built for that reader. They render in a browser, they look fine to you, and they hand a machine almost nothing it can quote. Some go further and block the very crawlers they most want to be seen by.
The scanner tells you which one you are, in about a minute.
Six things it grades.
Every check rolls up into one of these, and each category carries a different weight in the final score.
AI Search & Citation
The answer-engine layer, and the most heavily weighted part of the score. Whether ChatGPT, Claude, Perplexity and Gemini can understand your pages well enough to quote them.
Discoverability
The basics an AI crawler looks for first: can it find your pages at all, and does the site tell it where everything lives.
Content Accessibility
Whether your content survives the trip. A page that only assembles itself in a browser is a blank page to most of the things reading it.
Bot Access Control
What your site actually says to AI crawlers, rather than what you assume it says. Plenty of sites block the exact bots they want to be cited by.
Protocol Discovery
The newer standards agents use to work with a site instead of just reading it: llms.txt, schema.org, MCP, agent skills.
Commerce
For sites that sell, whether the things you sell are described in a form an agent can act on.
Your score becomes a level.
Each band says what an AI can actually do with your site, rather than handing you a grade to interpret.
| Level | Score | What it means |
|---|---|---|
| Invisible | 0-20 | AI can barely see the site. |
| Detectable | 21-40 | Findable, and little more. |
| Readable | 41-60 | AI can read the content. |
| Citable | 61-80 | Answer engines can trust it and quote it. |
| Agent-Ready | 81-100 | Fully usable by autonomous agents. |
The fixes are the point.
A score on its own is a diagnosis you cannot act on. So every failing check comes with a self-contained instruction written for a coding agent. Paste it into Claude Code, Cursor or Windsurf and the agent makes the change in your codebase.
Copy all hands over every prompt at once, including the optional ones that do not affect your score, each tagged required, improve or optional so the agent knows what is mandatory and what is not.
If your CDN challenges the scanner, it says so instead of grading the challenge page, and gives you a prompt for whoever manages that CDN.
Free, and no account.
Enter a URL and read the report. If you would rather not hand the prompts to an agent yourself, our AEO service does the whole job for you.