Free llms.txt generator and checker: enter your URL to build a valid llms.txt from your sitemap, or fill it in by hand, then validate any site's live llms.txt against the format before AI tools read it.
Generate an llms.txt
llms.txt
Validate an llms.txt
What an AI tool reads
Checks follow the llms.txt proposal at llmstxt.org. It is a voluntary format: each AI tool decides whether and how to read it.
How to make an llms.txt file
01
Name it and sum it up
Enter your site name and one sentence on what it does. Those become the H1 and the summary blockquote.
02
List your key pages
Add sections such as Docs, Guides or Products, one link per line. Put nice-to-have pages under Optional.
03
Download, upload, validate
Download the file, upload it to your site root as /llms.txt, then fetch it in the validator to confirm it is live and valid.
What llms.txt can and cannot do
An llms.txt file is a short markdown index of your site written for language models: your name, a one-line summary, and links to the pages that matter, each with a note on what it covers. The idea, proposed at llmstxt.org in 2024, is that a model reading one small file gets a cleaner picture of your site than it would by crawling menus, cookie banners and scripts.
Be clear about its limits. Google has said it does not use llms.txt for Search, and no major AI assistant has committed to reading it for everyday answers. Where it does get used is developer documentation and coding tools. It costs a few minutes to make, so it is worth having, but it will not replace crawlable pages, clear headings and content AI Overviews can quote.
llms.txt also cannot keep anything out. Access for AI crawlers is set in robots.txt, which you can test against GPTBot, ClaudeBot, OAI-SearchBot and others with the robots.txt tester. Both tools run in your browser; the optional fetch reads public files through a small server function and stores nothing.
Common questions
llms.txt is a proposed standard, published by Jeremy Howard of Answer.AI in September 2024, for a markdown file at the root of a website (example.com/llms.txt) that gives AI tools a short, curated map of the site: its name, a one-line summary, and links to its most useful pages grouped under headings. It is meant to help language models find clean content without parsing navigation, ads and scripts.
Markdown, in a fixed order. An H1 with the site or project name comes first and is the only required part. Then an optional blockquote summary, optional paragraphs with no headings, and any number of H2 sections, each containing a list of links written as - [Page title](https://url): optional note. A section headed Optional holds links a tool can skip when it is short on space.
Yes. Enter your site under Build from your site and it finds your sitemap (from robots.txt, or /sitemap.xml), then reads the title and meta description of up to 60 pages, starting with the pages your homepage links to. It groups them into sections, leaves out duplicates, logins, archives and legal pages, and flags missing or repeated descriptions. Everything lands in the form so you can edit it before you copy the file. By default it adds one credit link to PixelSERP under ## Optional. Untick the box and it is gone.
Google has said it does not use llms.txt for Search, and it has no effect on rankings. Some AI tools and coding assistants do read it, especially for developer documentation. Treat it as a cheap, low-risk addition rather than a ranking factor: it takes minutes to make with the generator above.
At the root of your domain, so it loads at https://yoursite.com/llms.txt, served as plain text. On most hosts you upload it next to robots.txt. Site builders that do not allow root files often serve their homepage instead, which the validator detects and flags.
llms.txt is the index: a short list of your key pages with one-line notes. llms-full.txt is an optional companion that holds the full text of those pages in one markdown file, so a tool can read everything in a single request. Keep llms.txt small and put long content in llms-full.txt.
No. robots.txt tells crawlers which URLs they may request; llms.txt suggests which pages are worth reading and summarises them. llms.txt cannot block anything. To control which AI crawlers can access your site, use robots.txt, and check it with the robots.txt tester.
That the file starts with an H1, has one H1 only, includes a summary blockquote, groups links under H2 sections, writes every list item as a markdown link, and uses full URLs. When you fetch a site it also checks the HTTP status and content type, so a missing file or an HTML page served in its place gets caught.
No. The generator and validator run in your browser. The only network requests are the optional build and fetch, which read a site's public sitemap, page titles and descriptions, or llms.txt through a small server function, because browsers cannot read another site's files directly. PixelSERP does not save your file or the URLs you check; Cloudflare, which runs that function, keeps standard short-lived request logs.