Skip to content
Scority
Start free audit

Free · no signup · ready to publish

llms.txt Generator

llms.txt is the emerging convention that tells AI assistants and dev-agents what your site is and which pages matter. Enter your site — we read your public sitemap and homepage metadata and build a sectioned, spec-valid file you can upload to your site root as /llms.txt.

We read your public robots.txt, sitemap and homepage metadata once. The URL isn't sent to marketing analytics.

How it works

Generation is deterministic — no AI model writes anything, so nothing is invented:

  • Sitemap discovery — we read robots.txt for Sitemap: lines, then fall back to the conventional /sitemap.xml. Sitemap indexes are expanded through their first child sitemaps.
  • Sections from your URL structure — pages are grouped by their first path segment (/docs/… → “Docs”), with readable titles derived from each URL slug.
  • Branding from your homepage — the H1 comes from your <title>and the summary blockquote from your meta description; if the homepage can't be read we fall back to the hostname.
  • Validation — every output is checked against the llmstxt.org structure (H1 first, well-formed - [title](url) items) before we show it.

What llms.txt is — honestly

llms.txt (see llmstxt.org) is a curated, Markdown-formatted index of your site for AI systems — the AI-era cousin of a sitemap. It is an emerging convention: some AI dev-tools and crawlers read it today, adoption is growing, and Google has stated it is not a ranking signal. Treat it as low-cost content infrastructure: cheap to publish, useful to the agents that do read it, and harmless otherwise.

Publishing the file

Save the generated file as llms.txt and upload it to your site root so it is reachable at https://yourdomain.com/llms.txt — next to robots.txt, not inside it. Review the titles first and remove anything you would not want AI tools to emphasise; the file is yours to curate.

Limitations

  • Sites without a reachable sitemap get a minimal starter file — publish a sitemap and regenerate, or add pages by hand.
  • Very large sitemaps are truncated to the first few hundred URLs.
  • Titles are derived from URL slugs, not fetched from every page — polish them by hand where it matters.
  • The generator reads only public files (robots.txt, sitemap, homepage) from our servers — password-protected or geo-fenced content is invisible to it.

Already have an llms.txt? Check whether AI crawlers can actually reach your site with the AI Crawler Access Checker, or run the Page SEO Checker for the full technical picture.