← All articles

Guides5 min read

llms.txt explained: an example, how to publish it and how to check it

What an llms.txt file is, a worked example line by line, how to publish one on WordPress or Shopify, whether any AI engine reads it, and how to check yours is real.

By Citeflare Team

An llms.txt file is a plain-text map of your site, written in Markdown, that sits at /llms.txt and lists your most useful pages for AI assistants. It is cheap to add. But no major AI engine has said it reads the file, and Google has said it does not use it. Add one if you want a tidy summary of your site that you control. Do not expect it to change where you are cited.

This guide covers the format, a worked example, how to publish it, what the evidence says, and how to check that yours is a real file and not an error page in disguise.

What llms.txt is

The proposal comes from llmstxt.org. The idea: a model has a limited reading window, and most websites are full of navigation, scripts and ads. A short Markdown file that points to your best pages is easier to read than a crawl of the whole site.

The format has a fixed order:

  1. An H1 heading with the name of the site or project. This is the only required part.
  2. A blockquote with a short summary of what the site is.
  3. Optional paragraphs with more detail, in any Markdown except headings.
  4. Optional H2 sections, each holding a list of links. Each item is a link, and can be followed by a colon and a note about the page.
  5. By convention, an H2 section named Optional. Links under it are the ones a reader can skip when short on space.

Because it is Markdown, people and machines can both read it.

An example, line by line

Here is a small file for a made-up bakery site.

# Rye & Rise Bakery

> Rye & Rise is a family bakery in Leeds. We sell sourdough, pastries and wedding cakes, and ship cake kits across the UK.

Prices are in pounds. Orders close at noon for next-day collection.

## Shop

- [Sourdough](https://ryeandrise.example/sourdough): loaves, with weekly availability
- [Cake kits](https://ryeandrise.example/kits): bake-at-home kits and delivery times

## Help

- [Delivery and returns](https://ryeandrise.example/delivery): areas, costs and cut-off times
- [Wedding cakes](https://ryeandrise.example/weddings): how to order, with lead times

## Optional

- [Our story](https://ryeandrise.example/about)

What each part does:

  • # Rye & Rise Bakery is the name. One H1, at the top.
  • The > line is the summary. Say what you are and what you sell, in plain words.
  • The paragraph under it holds facts a reader needs before following links, such as currency or order rules.
  • ## Shop and ## Help group links by purpose. Use names a visitor would use.
  • Each - [Title](url): note line is one page. The note says what the page is for.
  • ## Optional holds pages that matter less.

For a real example, Citeflare serves its own at citeflare.ai/llms.txt. It has a name, a summary, then sections for product, platform, solutions and blog posts, and an Optional section at the end.

How to write and publish one

  1. Pick 10 to 30 pages that answer the questions people ask about you: what you sell, pricing, help, key guides. Leave out login pages, tag archives and thin pages.
  2. Write the file in the format above. Keep notes short and factual.
  3. Publish the main file at the root of your domain, so it opens at https://yourdomain.com/llms.txt. The proposal also allows a file under a subpath, such as /docs/llms.txt, that covers just that section of the site.
  4. Serve it as plain text (text/plain) or Markdown, not as an HTML page.
  5. Update it when you add or remove important pages.

WordPress

On a standard WordPress install, a file you place in the site root (the folder that holds wp-admin and wp-content) is served as-is by the web server, so uploading llms.txt there through your host's file manager or SFTP works. Some SEO plugins also generate the file. Check your plugin's settings and, once it is live, open the URL to confirm what it serves.

Shopify

Since May 2026, Shopify's Online Store serves /llms.txt itself. By default it mirrors /agents.md, which holds instructions for AI shopping agents, not a page list in the llmstxt.org layout. To publish your own, add a templates/llms.txt.liquid file under Online Store, Themes, Edit code. Shopify looks for llms.txt.liquid first, then agents.md.liquid, then its own default. Only the request and agents Liquid objects are available in that template, so objects like shop and collections are not, and you keep the content current by hand. See Shopify's llms.txt.liquid documentation. Shopify's docs do not cover headless stores such as Hydrogen; there you need to serve the file yourself, for example through an app, a proxy or a Cloudflare Worker.

Does any AI engine use it?

No major engine has said it does.

  • Google Search ignores it. Google's AI optimization guide says you don't need to create "new machine readable files, AI text files, markup, or Markdown to appear in Google Search", and that creating llms.txt files "will neither harm nor help your site's visibility or rankings in Google Search". Google's John Mueller put it plainly on Bluesky in June 2025: "FWIW no AI system currently uses llms.txt."
  • OpenAI, Anthropic and Perplexity document their crawlers, as covered in AI crawlers explained. We found no commitment from any of them to read llms.txt. Check their current documentation, because this can change.

So treat llms.txt as a low-cost extra. It is quick to write. It does not replace the things that matter more: crawlers that can reach your pages, text in the HTML, and content worth citing. Citeflare rates a missing llms.txt as a low-severity item for that reason. We make no claim that having one raises citations or rankings, and you should be wary of anyone who does.

How to check yours is real

The common failure is not a missing file. It is a file that looks present but is not. Many sites answer every unknown URL with their normal HTML page and a 200 status code. This is called a soft 404. Your /llms.txt then "exists" but holds a web page.

Run this in a terminal:

curl -iL https://yourdomain.com/llms.txt

Look for three things:

  • The status is 200.
  • The Content-Type header is text/plain or text/markdown, not text/html.
  • The body starts with # Your Name, not <!doctype html> or <html.

If you see HTML, your server or CMS is catching the request. Fix the routing so the file is served directly.

llms.txt versus robots.txt and sitemaps

File Job Who it speaks to
robots.txt Says which crawlers may fetch which paths Crawlers, enforced by their good behaviour
sitemap.xml Lists every URL you want found, with dates Search engine crawlers
llms.txt Summarises and curates your best pages in readable text Any reader, human or model

robots.txt controls access. A sitemap is a complete list. llms.txt is a short, hand-picked one. If robots.txt blocks a crawler, an llms.txt file does not let it back in. For the crawler side, read AI crawlers explained.

How Citeflare checks and generates it

Citeflare's AI readiness audit requests https://<your-domain>/llms.txt. It counts the file only if the response is a success, the body is not empty, the body does not start with <, and the content type is not HTML. A soft 404 fails the check, and you get a low-severity action: "Publish an llms.txt file".

To fix it, Citeflare builds a file from the pages in your crawled sitemap that returned a 200 status. It uses your brand name as the heading, your project description as the summary, and one link per page, all under a single ## Pages heading with no notes. You download it as llms.txt, then edit the summary, drop pages that do not help a reader, add a short note to each link, group the pages into sections, and upload it to your site root. It needs a crawled sitemap first, so crawl it before you download.

To run the audit, get started, add your site, and open the AI readiness tab on the Sitemap page. To see which AI crawlers actually visit your pages, read about crawler analytics or see the full platform overview. For the bigger picture, see the generative engine optimization guide.

FAQ

What is an llms.txt file?

A Markdown file at /llms.txt that names your site, summarises it and lists your key pages with short notes.

Is there an llms.txt example I can copy?

Yes, the bakery file above shows every part. citeflare.ai/llms.txt is a real one.

Does Google use llms.txt?

Google Search ignores it. Google says you do not need AI text files to appear in Google Search, and John Mueller said in June 2025 that no AI system currently uses it.

Is there an llms.txt checker?

You can check by hand with curl -iL as shown above. Citeflare's AI readiness audit runs the same check, including the soft 404 test.

Is there an llms.txt file generator?

Citeflare builds one from your crawled sitemap pages for you to edit and download. Some SEO plugins and other tools also generate them. Review whatever they produce before you publish.

How do I add llms.txt to Shopify?

Shopify's Online Store serves a default /llms.txt itself. To replace it, add a templates/llms.txt.liquid file to your theme. Headless stores need to serve the file themselves.

Will llms.txt get me cited in AI answers?

We have no evidence that it does, and no major engine has said it reads the file. Fix crawler access and page content first.

See what AI says about you. Your first scan runs the day you sign up.

Monthly or annual plans. Cancel any time.