AI website analyzer: is your site ready for AI?

How to use the AI readiness grader

1. Enter your website. The grader checks the home page of that host and the files at its root.
2. Read the score, then the list. Checks that lost points come first, ordered by the points you can win back.
3. Fix the top item and grade again. Most checks link to the free tool that fixes them. Results are cached for one hour, so wait an hour or check another host in between.

The 8 checks and their points

A check that partly passes gets half its points, or a share for crawler access (3 of 4 AI crawlers allowed gives 15 of 20). The grade labels are: 85 and up Ready, 60 to 84 Mostly ready, 35 to 59 Needs work, under 35 Not ready.

Why these checks

AI assistants answer from pages their crawlers can fetch and read. Three things stop that most often, and they carry the most points:

• A robots.txt rule that blocks the crawler. Many sites block AI crawlers by copying a list from the internet, without separating training crawlers from search crawlers. The grader tests only the search and user-fetch crawlers, because those decide whether an assistant can cite and link you today.
• Text that only appears after JavaScript runs. Many crawlers read the HTML as it arrives. A single-page app that renders in the browser looks empty to them.
• No clear help or docs pages. Questions about a product are answered from its help center, docs and FAQs. If those pages are not linked or listed, crawlers and support chatbots may never find them.

The other checks make the site easier to understand: a sitemap so every page is found, structured data that states what the business is, an llms.txt file that points AI tools to the pages that matter, and a title and description that summarize the page.

What to fix first

1. Crawler access. Test your rules with the robots.txt checker. If you want to block model training, block the training tokens and keep the search tokens allowed.
2. Server-rendered text. Ask your developer to use server-side rendering or static generation for marketing, docs and help pages.
3. A sitemap that works. Check it with the sitemap checker and list it in robots.txt.
4. Findable help content. Link your help center or docs from the main navigation or the footer.
5. llms.txt and structured data. Draft an llms.txt with the llms.txt generator and add Organization JSON-LD to the home page.

What the grader does not measure

The score is about whether machines can read your site, not whether your content is good. It does not check rankings, backlinks, page speed or whether your pages answer real questions clearly. It reads only the home page and the root files of the host you enter, so a docs site on its own subdomain needs its own check.

A readable site is the first step for a support chatbot too. To see how a help center becomes a chatbot that answers from those pages, read how to turn your help center into an AI chatbot.