LLMs.txt Generator
Generate a spec-compliant llms.txt file for your site. AI models read this file to understand your structure — get yours up before competitors do.
Why LLMs.txt Generator matters
llms.txt is to AI what robots.txt is to search engines: a structured map that tells a model what your site contains and which pages matter. Sites without one leave the model to infer your hierarchy from navigation markup and guess wrong. The standard is young but already adopted by Anthropic, Mistral, Cursor, and Stripe, and writing one by hand is an afternoon of copy-pasting URLs you already have.
How it works
- 01
We look for your sitemap
Enter your domain and we try the four conventional sitemap locations. If one answers we take your own declared URLs, which is far more reliable than crawling your navigation.
- 02
We group the pages into sections
URLs are bucketed by path into Product, Documentation, Blog, Tools, Company, and Other, up to fifty pages. That grouping is the part of the spec that actually helps a model, and the part people skip when writing one by hand.
- 03
You copy or download the file
The output is spec-compliant markdown with a title, a summary line, and grouped links. Review it, edit anything that reads wrong, and upload it to /llms.txt at your root.
What it checks
- Sitemap discovery
- Tries /sitemap.xml, /sitemap_index.xml, /sitemap-index.xml, and /sitemap before falling back to reading your homepage links.
- Section bucketing
- Path-pattern matching sorts pages into the sections the spec expects rather than dumping one flat list.
- Page summaries
- Each entry carries a short blurb pulled from the page itself, so a model gets a reason to follow the link.
- Spec-compliant output
- H1 title, blockquote summary, H2 sections, markdown link lists. Formatting the file wrong is the most common reason one gets ignored.