llms.txt Generator: Build an llms.txt From Your Sitemap

Paste a URL list or your sitemap.xml and get a ready-to-upload llms.txt that follows the llmstxt.org proposal, with sections grouped by folder and low-priority links moved to Optional.

Last updatedHow we build & check our tools
Results will appear here once you enter your values.

How the generator works

Paste URLs in any mix of formats: one URL per line, a Title, URL pair, a Markdown link such as [API reference](https://example.com/api): endpoints and errors, or the full XML of a sitemap.xml. The tool reads every loc tag or link, drops duplicates and fragments, and counts lines without a valid http or https URL. Nothing is fetched, so it works on staging sites and private docs.

  • Header: your site name becomes the H1 (it defaults to the hostname of the first URL), the summary becomes a blockquote, and extra details follow as plain Markdown.
  • Sections: links are grouped by the first folder in the path, so /docs/quickstart lands under Docs and root pages such as /pricing land under Pages. You can rename, drop or move any group to Optional.
  • Titles: when no title is given, one is built from the URL slug, so /docs/getting-started becomes Getting Started.

Copy the result or download it as llms.txt.

What the llms.txt proposal specifies

llms.txt was proposed by Jeremy Howard in September 2024 on llmstxt.org. It is a Markdown file served at /llms.txt that gives language models a short, curated map of a site instead of making them parse navigation, ads and scripts. The order is fixed: an H1 with the project name (the only required part), a blockquote summary, optional detail paragraphs without headings, then H2 sections containing lists of links written as a Markdown link followed by optional notes after a colon.

A section named Optional has a special meaning: its links are secondary and can be skipped when an agent's context is short. The generator puts blog, news, changelog, tag, category, author, archive, press, legal, privacy and terms folders there by default.

The proposal also suggests publishing clean Markdown copies of pages at the same URL plus .md. Enable the Markdown option and page links become page.md, and folder URLs ending in a slash become index.html.md.

Common mistakes

Expecting an SEO boost. llms.txt is a community proposal, not a standard that search engines or AI crawlers must read. Google's Search Central guide says Google Search ignores such files and they neither help nor harm rankings. Its value is for agents, IDE assistants and tools that choose to fetch it.

Pasting a sitemap index. A sitemapindex file lists other sitemaps, not pages. The tool detects this and warns you; open each child sitemap and paste its XML instead.

Linking to Markdown that does not exist. Only turn on the .md option if your server actually returns those files, otherwise every link in the file is a 404.

Dumping the whole sitemap. A 5,000 URL list defeats the purpose of a short index. Use the exclude filters and the per-section cap.

Tips for a useful llms.txt

Write the summary for a model that knows nothing about you: what the product is, who it is for, and what it should know first. Put facts that change answers, such as all prices are in USD or v1 of the API is deprecated, in the extra details box. Any headings you type there are converted to bold lines, because the proposal does not allow headings before the link sections.

  • Keep the default exclude patterns /tag/, /author/ and /page/ to strip archive and pagination URLs.
  • Cap busy folders at 10, 25, 50 or 100 links so no single folder crowds out the rest of the file.
  • Add a note after each important link describing what the page covers.
  • Upload the file to your site root so it resolves at https://yourdomain.com/llms.txt, and regenerate it when your docs change.

Frequently Asked Questions

Common questions about the llms.txt Generator: Build an llms.txt From Your Sitemap

It is a Markdown file placed at the root of a website, at /llms.txt, that summarizes the site and links to its most useful pages for large language models. The format was proposed on llmstxt.org in September 2024. It contains an H1 name, a blockquote summary, optional notes, and H2 sections of annotated links, with an Optional section for pages that can be skipped.
From the same team

Stop paying per token — route AI requests to your own GPU

Wide Area AI is a local-first AI gateway: repeated requests hit an edge cache, the rest run free on your own hardware, and the cloud is only a failover. OpenAI-compatible endpoint, free tier.

Start routing — free

Sources & References

llmstxt.org: The /llms.txt file proposal

The llms.txt proposal: H1, blockquote, detail text, H2 file lists, Optional section, .md URLs.

Google Search Central: Optimizing for generative AI features on Google Search

Google Search does not use llms.txt.

sitemaps.org: Sitemaps XML format and sitemap index files

Sitemap XML <loc> format and sitemap index files.