What is llms.txt?

llms.txt is a plain Markdown file at the root of a website (yoursite.com/llms.txt) that gives AI tools a short, curated list of the site's most useful pages. Jeremy Howard proposed the format in September 2024. It's voluntary and quick to write. Google says it isn't needed for its AI features, and the major AI search products haven't said they use it. It's a small, optional addition, not a ranking tactic.

What goes in an llms.txt file?

The format is described at llmstxt.org. In order, a file has:

1. An H1 with the site or project name. This is the only required part.
2. A blockquote with a short summary.
3. Optional notes in plain Markdown, with no headings.
4. H2 sections of links. Each item is [Page name](URL), optionally followed by a colon and a short note.
5. An optional section named "Optional" for secondary links a tool can skip when it's short on space.

A shortened excerpt from the file GoPoint serves at gopoint.com/llms.txt:

# GoPoint

> GoPoint is an independent, senior-led digital growth studio in Calabasas, California, in business since 2012. We build websites, AI tools and content for marketing teams, publish our pricing, and measure what the work returns every month.

## Website builds
- [Website redesign services](https://gopoint.com/website-redesign-services): Redesigns planned around customer questions, with redirects and launch checks. Launch $15K–$30K; Replatform $30K–$80K; Platform $80K–$150K+. Scope depends on content, integrations and migration work.
- [Website migration services](https://gopoint.com/website-migration-services): Moves from WordPress, Squarespace, Wix, Webflow or Drupal with a URL inventory, redirect map and launch checks. Replatforms from $30K.

The proposal also suggests publishing a clean Markdown copy of important pages at the same URL with .md added, so a tool can read the content without navigation and scripts.

Where does llms.txt go?

At the root of your domain: https://yoursite.com/llms.txt. The proposal also allows a file at a subpath, such as /docs/llms.txt, covering the pages under it. Serve it as plain text with a 200 status, and make sure robots.txt doesn't block it.

To see another site's file, add /llms.txt to its domain in your browser, or enter the domain in the checker above.

Does llms.txt help SEO or AI citations?

There's no public evidence that it does.

• Google says you don't need new machine-readable files or "AI text files" to appear in AI Overviews or AI Mode (Google Search Central). Gary Illyes said in July 2025 that Google doesn't support llms.txt (Search Engine Land).
• OpenAI's and Anthropic's crawler documentation describes how their bots use robots.txt and doesn't mention llms.txt as a signal (OpenAI, Anthropic).
• Server logs suggest the major AI crawlers request it rarely. In one 12-week study of 83 sites (April–July 2026), OpenAI's crawlers requested llms.txt 7 times and robots.txt 3,990 times. PerplexityBot didn't request it at all. Meta's crawler requested it more often than robots.txt (EZY Research). EZY sells llms.txt tools, and its sample skews toward small-business sites.

What does affect AI search is the groundwork:

• whether the right crawlers can reach your pages;
• whether your pages show their content without JavaScript;
• whether the answers on the page are clear;
• whether your business details are consistent everywhere they appear.

The AI Search Readiness Check reviews your site's technical readiness on several of these.

Training crawlers vs AI search crawlers

AI companies run separate crawlers for training models and for finding pages to cite in search answers. Blocking one doesn't block the other.

So you can block GPTBot to opt out of OpenAI training while still allowing OAI-SearchBot, so your pages can appear in ChatGPT search. Google says Google-Extended doesn't affect inclusion or ranking in Google Search.

User-requested fetchers such as ChatGPT-User and Perplexity-User act for a person asking a question, and the companies say robots.txt may not apply to them in the same way. An llms.txt file doesn't change any of these rules. robots.txt does.

Should I create an llms.txt file?

It's reasonable if your site has documentation, a product, or more than a handful of service pages, and someone will keep it current. Some coding assistants and agent tools do read it. It's lower priority than your robots.txt, sitemap and page content, so sort those first.

If nobody will maintain it, skip it. A file listing old prices or dead pages gives tools wrong information.

How to create an llms.txt file

1. Pick 10–30 pages a buyer or user would actually need: products or services, pricing, documentation, support, about and contact.
2. Write the H1 and a one-sentence summary: what you do, for whom, and where.
3. Group the links under H2s that match how your site is organized, with a short note after each.
4. Put nice-to-have pages under "## Optional".
5. Publish it as llms.txt at your site root (most CMSs and hosts let you upload a static file) and confirm it opens at /llms.txt.
6. Check it with the tool above, and update it whenever prices, services or URLs change.

Leave out anything private: internal tools, staging links, customer data or unpublished pricing.

llms.txt vs robots.txt vs sitemap.xml