llms.txt is a lightweight Markdown file that gives AI assistants a fast, structured way to understand what your site is about — without having to crawl and parse your full HTML. It sits at the root of your domain, follows a simple format (an H1 title, a blockquote summary, and H2 sections of links), and is meant to be read directly by language models rather than rendered in a browser.
Why it matters for AI search: ChatGPT, Claude, Perplexity, and similar tools increasingly browse the live web to answer questions and cite sources. A well-structured llms.txt gives these systems a concise, unambiguous summary instead of forcing them to infer your site's purpose from navigation menus, ads, and boilerplate. That can make the difference between an AI assistant accurately describing your product and citing the right page, versus skipping your site entirely in favor of one that's easier to parse.
How this tool works: fill in your site name and a one-sentence summary — these become the required H1 and blockquote at the top of the file. Add any extra context AI assistants should know. Then build out H2 sections (Docs, Product, Blog, etc.) and add links with optional short descriptions. You can also paste a sitemap.xml URL and the tool will fetch it through our proxy and prefill a "Pages" section from the URLs it finds, which you can then edit down to the pages that actually matter. The output updates live, and you can copy it to your clipboard or download it directly as llms.txt.
Limitations: this tool only generates the file — you still need to upload it to your site's root directory yourself (via your CMS, static site host, or web server config). It doesn't validate an existing llms.txt file against the spec; use our llms.txt Validator for that. The sitemap prefill feature is a convenience shortcut, not a substitute for manually curating which pages actually deserve to be listed — a shorter, well-chosen list of links is more useful to an AI assistant than an exhaustive dump of every URL on your site.