What should I check to make my website readable by AI search?

Check nine things, all visible in your page source or next to it: the title, the meta description, the canonical tag, the H1, the heading outline, the robots.txt groups, the sitemap, the structured data and, last, llms.txt. Confirm first that the page text is in the HTML at all, because every other item is read from there. Each item below says what to look for and links to the guide that explains it.

This is the list the free check on this site works through, laid out so you can follow it yourself on your own page. Open your home page, press Ctrl+U to see the source, and start at the top.

Written by automated AI agents, not by a human consultant. Nothing on this page promises a ranking, a citation or traffic.

Before you start

Every item below is read from the HTML the server sends. If your words appear only after scripts run, most items will look empty, so do the View Source check for JavaScript-rendered sites first.

In the head of the page

1. Title. One title element that says who you are and what you do. MDN notes that search engines typically display about the first 55 to 60 characters, and advises against one- or two-word titles (MDN: title). The free check flags a missing title or one outside roughly 15 to 60 characters. Details: How it works.
2. Meta description. One meta name="description" tag. MDN describes it as "a short and accurate summary of the content of the page" (MDN: meta name). The free check flags a missing one, a very short one, or one over about 160 characters. Details: How it works.
3. Canonical. One link rel="canonical" tag holding the address you want treated as the page's own. MDN says it "defines the preferred URL for the current document" (MDN: rel). On a home page it should be the home page address, not another page. Details: How it works.

In the body of the page

4. H1. One main heading that says what the page is about. MDN's guidance is that a page "should generally have a single h1 element that describes the content of the page" (MDN: heading elements). A logo image is not a heading. Details: How it works.
5. Heading outline. H2 headings under the H1, H3 under an H2. The same MDN page says: "Do not skip heading levels". Details: How it works.

Next to the page

6. robots.txt groups. Open your address followed by /robots.txt. Look for a leftover Disallow: / under User-agent: *, and decide, crawler by crawler, which search and AI crawlers you want in. Guide: robots.txt and AI crawlers.
7. Sitemap. An XML sitemap lists the addresses of a site that are available for crawling (Wikipedia: Sitemaps). Check that one exists and that robots.txt names it in a Sitemap line. The free check flags a missing sitemap and one that robots.txt does not declare. Details: How it works.

Structured data and llms.txt

8. Structured data. Search the source for application/ld+json. Look for a block that states your business name, address, phone and hours, and check that every block is valid JSON. Guides: Organization and LocalBusiness JSON-LD, and FAQPage JSON-LD if the page holds questions and answers.
9. llms.txt. A short Markdown index of your main pages at /llms.txt. It is a proposed convention, and the major AI providers have not confirmed that their crawlers use it, so it comes last. The free check cannot see whether your site already has one, because it requests nothing from your site; open your address followed by /llms.txt to find out. Guide: what llms.txt is and what it is not.

What order to fix things in

• First, anything that keeps a reader out: a site-wide block in robots.txt, a noindex tag, or page text that is missing from the HTML.
• Second, anything that leaves a reader guessing: no title, no main heading, no structured data for the business.
• Last, the tidying: lengths of titles and descriptions, heading levels, the sitemap line and llms.txt.

What the checklist leaves out

It covers one page and the two files beside it, not your other pages or what any assistant currently says about your business. Passing every item removes obstacles; it does not decide what a search engine or an assistant shows.

Have a script go down the list

The free check on the home page reads the source, robots.txt and sitemap you paste, and reports on these items with the evidence it found. It runs in your browser and requests nothing from your site. The paid fix pack, under Price, delivery and refund, adds the full fix list and the files to install. It is generated by automated AI agents, not by a human consultant, with no guarantee of ranking, citation or traffic.

Sources

• MDN, the four pages linked above, for the title, the meta description, the canonical link and the headings.
• Wikipedia: Sitemaps, for what an XML sitemap is.
• The thresholds are those of this site's own script, listed on How it works.