llms.txt file: what it is, format and example

An llms.txt file is a short Markdown file at /llms.txt that tells AI agents what your site is and which pages matter most. It was proposed by Jeremy Howard in September 2024 and is documented at llmstxt.org. It is a proposal, not an official standard, and Google says Google Search does not use it.

What is llms.txt?

llms.txt is one plain Markdown file that gives AI agents a shortcut to the facts on your site.

Web pages are built for people. They carry menus, scripts, cookie banners and ads. An AI agent that only needs the facts has to dig through all of that.

llms.txt gives the agent a shortcut. It is one plain file with:

the name of the site,
a one-paragraph summary,
a curated list of links to the pages that answer real questions.

The idea comes from Jeremy Howard. He published the proposal on 3 September 2024, on llmstxt.org and on the Answer.AI blog. Version 2 of the proposal was published on llmstxt.org in August 2026.

The proposal says the file is mainly meant for inference, not training. In plain words: it helps an agent that is answering a question right now. It is not a request to be included in a training dataset.

llms.txt is not robots.txt

People often mix these two up. They do different jobs.

File	Job	Who reads it
`robots.txt`	Says which bots may crawl which paths	Crawlers, before they fetch pages
`sitemap.xml`	Lists every indexable page	Search engines
`llms.txt`	A short, curated guide to the most useful content	AI agents, on demand

llms.txt does not allow or block anything. If you want to control which AI bots can fetch your pages, that is robots.txt. See our AI crawlers guide.

What format does llms.txt use?

llms.txt uses Markdown, with a fixed order of sections defined by the llmstxt.org proposal. Markdown means people and language models can both read it. It is also strict enough to parse with simple code.

An H1 with the site or project name. This is the only required part.
A blockquote with a short summary. Put the key facts a reader needs to understand the rest of the file here.
Optional free text below the summary: paragraphs or lists, but no headings.
Zero or more H2 sections, each holding a “file list”. A file list is a Markdown list. Each item is a link, [name](url), optionally followed by a colon and a short note.

By convention, an H2 called ## Optional holds secondary links. An agent can skip them when it needs a shorter context. Since v2 the name has no special meaning; it is just a convention.

The proposal also suggests a clean Markdown version of each important page at the same URL plus .md. For example, page.html.md, or page.md with the extension replaced. Version 2 adds two link relations so agents can find these files without guessing:

rel="alternate" type="text/markdown" points from a page to its Markdown version.
rel="describedby" points from a page to the llms.txt file that covers it.

These can be HTML <link> elements or an HTTP Link: header.

A minimal llms.txt example

Here is a small, valid file for a made-up bakery. It follows the order above.

# Example Bakery

> Example Bakery is a family bakery in Springfield. We bake bread and cakes daily and take custom cake orders with 48 hours' notice.

Opening hours and prices change by season. Always link to the pages below rather than repeating them.

## Main pages

- [Menu](https://example.com/menu.md): Breads, pastries and cakes we bake every day
- [Custom cakes](https://example.com/custom-cakes.md): How to order, lead times and sizes
- [Visit us](https://example.com/visit.md): Address, opening hours and parking

## Policies

- [Allergens](https://example.com/allergens.md): Which products contain nuts, gluten or dairy
- [Refunds](https://example.com/refunds.md): Cancellations and refunds for custom orders

## Optional

- [Our story](https://example.com/about.md): How the bakery started

For a real-world file, llmstxt.org shows a cut-down version of the one used by the FastHTML project. It starts like this:

# FastHTML

> FastHTML is a python library which brings together Starlette, Uvicorn, HTMX, and fastcore's `FT` "FastTags" into a library for creating server-rendered hypermedia applications.

It then lists ## Docs, ## Examples and ## Optional sections of links.

Where do you put llms.txt?

Put the file at the root of your site, so it loads at https://example.com/llms.txt. We recommend serving it as plain text.

Version 2 of the proposal also allows a file at a subpath, such as /docs/llms.txt. That file covers the pages under /docs/. If more than one file applies, agents should use the most specific one. This helps people who only control one folder on a shared host.

Some documentation platforms also serve the file under /.well-known/. Mintlify, for example, serves it at both /llms.txt and /.well-known/llms.txt. The llmstxt.org proposal itself uses the root or a subpath.

llms-full.txt

You will often see a second file called llms-full.txt. It is not part of the llmstxt.org proposal.

It is a convention from documentation tools. Mintlify, for example, describes its llms-full.txt as a file that “combines your entire documentation site into a single file as context for AI tools and LLM indexing.”

The difference is simple:

llms.txt is a short map. It links out to detail.
llms-full.txt is the whole book in one file.

A full file can be very large. For a small site it adds little. For big documentation sets, some teams find it handy for pasting into a coding tool. Publish it only if you can keep it up to date.

Does anyone read llms.txt?

Google Search does not: Google says “Google Search itself doesn’t use them,” and we found no primary statement from OpenAI, Anthropic or Perplexity that their assistants read it.

llmstxt.org describes itself as “a proposal to standardise on using an /llms.txt file”. It is maintained on GitHub and open for community input. No standards body has adopted it.

Does Google use llms.txt?

Google Search does not use llms.txt. Google’s guide to optimizing for generative AI features (last updated 10 July 2026) says:

You don’t need to create new machine readable files, AI text files, markup, or Markdown to appear in Google Search (including its generative AI capabilities), as Google Search itself doesn’t use them.

The same guide adds that it is “completely fine” to keep llms.txt files for other services, and that “doing so will neither harm nor help your site’s visibility or rankings in Google Search, as Google Search ignores them.”

Do ChatGPT, Claude and Perplexity use llms.txt?

We found no primary statement from OpenAI, Anthropic or Perplexity saying their crawlers or assistants use llms.txt when they answer questions. llmstxt.org notes that OpenAI, Anthropic and Google’s Gemini publish llms.txt files for their developer documentation. Publishing a file is not the same as reading other people’s files. If such a statement appears, we will add it here.

Does Lighthouse check llms.txt?

Yes. Lighthouse has an Agentic Browsing category. Its docs list an llms.txt check that “checks for the presence of a machine-readable summary at the domain root.” Chrome says the Agentic Browsing category is experimental and based on proposed standards. So, inside Google, Search ignores the file while a browser audit looks for it. These are two different products with two different goals.

Platform support: WordPress, Shopify, Wix and others

Many tools can generate the file for you. Only claims we could source are listed.

WordPress (Yoast SEO). Yoast’s help pages explain how to switch llms.txt on in the plugin’s site features settings.
Shopify (Yoast SEO for Shopify). Yoast announced an opt-in llms.txt feature for its Shopify app on 31 March 2026. In automatic mode it rebuilds the file weekly from store data, including top-selling products, collections, policies and recent posts. We found no Shopify help article that documents a native llms.txt setting, so we do not claim one.
Docs platforms. Mintlify generates both llms.txt and llms-full.txt for sites it hosts. llmstxt.org also lists GitBook, Wix, the AIOSEO WordPress plugin, a Drupal recipe, and plugins for VitePress and Docusaurus.

If your platform is not listed, you can write the file by hand. It is just text.

llms.txt best practices

Keep it small, accurate and boring. The proposal’s own advice is to “use concise, clear language” and to test the file by asking an agent questions with only your llms.txt as a starting point.

Lead with a factual summary. Say what you do, for whom, and where. One short paragraph.
Curate, do not dump. Link only the pages that answer real questions. Your sitemap already lists everything.
Write a useful note on each link. “Refunds for custom orders” beats “Click here”.
Link to clean content. Markdown versions of pages are best. Plain HTML pages are fine if they are light.
Use the Optional section for nice-to-have pages.
Keep it true. Do not put claims in llms.txt that are not on your site. Agents may quote it.
Update it when pages move. Broken links waste the agent’s time.
Serve it as plain text with a 200 status. No login, no redirect chain, no bot challenge.
Do not expect rankings from it. Google says it has no effect on Google Search. Treat it as a help for agents, not an SEO lever.

llms.txt is one part of a site that agents can use. Robots.txt rules decide whether AI bots can fetch your pages at all. WebMCP goes further and lets an agent call actions on your page.

How to check your llms.txt

Open https://yourdomain.com/llms.txt in a browser. You should see plain Markdown text, not your home page and not an error page. Then check that:

the first line is an H1 (# Your site name),
a > summary follows,
every link loads.

You can also run our free agent-readiness check. It reports whether an llms.txt file was found on your site, along with your AI crawler access and WebMCP status.

For a deeper review of how agents and AI search see your site, Convertlyft offers an agent-readiness report.

Questions people ask

What is llms.txt?

llms.txt is a Markdown file placed at /llms.txt on a website. It gives AI agents a short summary of the site and a curated list of links to the most useful pages. It was proposed by Jeremy Howard in September 2024 and is described at llmstxt.org.

Is llms.txt an official web standard?

No. llmstxt.org calls it a proposal. It is not published by a standards body such as the IETF or W3C. It is open for community input on GitHub.

Does Google use llms.txt?

Not for Google Search. Google's guide to generative AI features says Google Search does not use these files, and that having one will neither harm nor help your visibility or rankings in Google Search. Separately, Chrome's Lighthouse has an experimental Agentic Browsing category that checks for llms.txt.

What is llms-full.txt?

llms-full.txt is a single file that contains the full text of a site's documentation. It is not part of the llmstxt.org proposal. It is a convention used by some documentation platforms, such as Mintlify.

Where do I put llms.txt?

At the root of your site, so it loads at https://example.com/llms.txt. Version 2 of the proposal also allows a file at a subpath, such as /docs/llms.txt, which then covers the pages under that path.

Sources

The /llms.txt file, v2 (llmstxt.org) (read 2026-10-24)
llms.txt changes: v2 (August 2026) (llmstxt.org) (read 2026-10-24)
/llms.txt — a proposal to provide information to help LLMs use websites (Answer.AI) (read 2026-10-24)
Google's guide to optimizing for generative AI features on Google Search (read 2026-10-24)
AI features and your website (Google Search Central) (read 2026-10-24)
Lighthouse agentic browsing scoring (Chrome for Developers) (read 2026-10-24)
llms.txt (Mintlify docs) (read 2026-10-24)
How to enable llms.txt with Yoast SEO (read 2026-10-24)
Yoast SEO for Shopify: llms.txt (31 March 2026) (read 2026-10-24)