What Is llms.txt and Does Your Site Need One?

llms.txt is a plain Markdown file that lives at the root of your website and gives AI systems a short, curated map of your most important pages. The format was proposed in September 2024, takes about an hour to adopt, and no major AI provider has officially confirmed reading it yet — here is what that means in practice.

We want to be honest with you from the first paragraph, because this little file has attracted more hype than almost anything else in AI search. So this guide does three things. It explains what llms.txt is in plain words. It shows exactly what the file looks like. And it lays out the real evidence — for and against — so you can decide in five minutes whether your site needs one.

What is llms.txt in simple terms?

llms.txt is a text file written in Markdown that sits at a fixed address: yoursite.com/llms.txt. The file opens with a one-sentence summary of what your site is about, then lists links to your most useful pages, each with a short description. Jeremy Howard, co-founder of Answer.AI, proposed the format in September 2024, and the full specification lives at llmstxt.org.

One picture makes it concrete. Your website is a house, and an AI assistant is a guest with very little time. Left alone, the guest wanders the hallways, opens a few random doors, and leaves with whatever it happened to see. llms.txt is the note you tape to the front door: welcome, the kitchen is here, the library is there, and this is the one room that matters.

The idea exists because of a real technical limit. When an AI model reads your site, everything passes through its working memory — the context window — and that memory is small compared to a whole website. Menus, cookie banners, and layout code eat the budget fast. A clean Markdown file is cheap to read, so the model can spend its attention on your actual content instead of your furniture.

What does an llms.txt file look like?

The format is deliberately simple, and that is its best feature. An llms.txt file has four parts: a Markdown H1 with the site name, a blockquote with a one-line summary, H2 sections that group related links, and the links themselves. Here is a complete, valid example for an imaginary bakery.

# Sunrise Bakery

> Sunrise Bakery is a family bakery in Austin, TX. We bake sourdough daily,
> deliver locally, and run weekend baking classes.

## Products

- [Sourdough menu](https://sunrise-bakery.com/breads.md): Daily breads and prices
- [Wedding cakes](https://sunrise-bakery.com/cakes.md): Custom orders and lead times

## Company

- [About us](https://sunrise-bakery.com/about.md): Our story and our bakers

## Optional

- [Press](https://sunrise-bakery.com/press.md): Reviews and awards

Two details are easy to miss. The Optional section is part of the specification: it marks links an assistant may skip when its reading budget is tight. And the links point to .md versions of pages where possible, because clean Markdown is far easier for a model to digest than rendered HTML.

One assumption underneath all of this is worth saying out loud: whatever page you link to has to carry its content in the HTML it returns. Most AI crawlers read that raw HTML and never execute JavaScript, so a page that looks perfect in your browser can still arrive empty.

What is llms.txt used for today?

Real usage in 2026 is narrower than the hype, but it is not zero. Three scenarios are worth knowing about:

• AI coding assistants use llms.txt the most: documentation sites publish the file so that AI-powered editors can pull a whole manual into the model in one clean request. Anthropic, the company behind Claude, maintains one for its own developer docs.
• Smaller AI agents and answer tools fetch llms.txt as a cheap way to understand an unfamiliar site before deciding which pages to read in full.
• Site owners use the file as a forcing exercise: choosing the ten links that explain your business best is valuable on its own — even if no robot ever reads the result.

How is llms.txt different from robots.txt and sitemap.xml?

One sentence clears up most of the confusion: robots.txt controls access, sitemap.xml lists everything for search engines, and llms.txt curates the best of your site for AI readers. The three files are neighbors, not competitors, and a healthy site can have all of them. The table below adds Schema.org markup, because all four standards get mixed up constantly.

The main takeaway from the table: llms.txt is the only one of the four that is still a proposal. robots.txt, sitemap.xml, and Schema.org are boring, confirmed infrastructure that every site should have. llms.txt is a bet — a small and cheap one, but a bet, and it deserves to be treated as one.

Of those three confirmed standards, Schema.org is the one that works inside your pages rather than beside them: a small JSON-LD block that states your name, your prices, and your FAQ answers in a vocabulary machines already parse. Our guide to schema markup and AI visibility shows which types are worth adding and what the code looks like.

One misconception is worth clearing up separately, because half the articles online repeat it: llms.txt does not block anything. The file cannot forbid AI companies to crawl your site or to train on your content — it is an invitation, not a fence. If you want to keep a crawler out, that job belongs to robots.txt and its user-agent rules for bots like GPTBot or ClaudeBot. llms.txt only says: if you do come in, read this first.

Does anyone actually read llms.txt?

Adoption is real. As of June 2026, 8.7% of the top 1,000 websites publish an llms.txt file, according to Rankability's monthly adoption tracker, and the share has been climbing all year. Serious engineering companies are on the list, and in the world of AI documentation the format is already treated as normal.

Reading is another story. No major AI provider — not OpenAI, not Anthropic, not Google — has officially announced that its assistant fetches llms.txt when answering. Google's John Mueller went further and compared the file to the old keywords meta tag: a place where a site describes itself, which search systems learned to ignore. And so far, no study has shown a measured link between having the file and being cited more often in AI answers.

Think of llms.txt as a porch light. It costs almost nothing to leave on, most guests will walk right past it, and the one guest who needs it will be very glad it was there.

That mix is why our advice lands in the middle. Do not pay anyone serious money for llms.txt magic, and do not expect the file to move your traffic this quarter. But the one hour it takes to publish an accurate file is a fair price for optionality: the format already works in AI coding tools today, and if a major assistant starts reading it tomorrow, your note is already taped to the door.

How do I create an llms.txt file?

You can write a solid llms.txt in under an hour, by hand, in any text editor. The steps:

1. Pick the five to ten pages that explain your business best — product pages, pricing, FAQ, and the about page are the usual suspects.
2. Write one honest sentence about what your site is and who it serves, and put it in the blockquote at the top.
3. Group the links under H2 sections, and give every link a short plain-language description after a colon.
4. Save the file as plain text named llms.txt and upload it to the root of your domain, right next to robots.txt.
5. Open yoursite.com/llms.txt in a browser to confirm it loads, and update the file whenever your key pages change.

Each of those steps has details worth getting right, and they live in our step-by-step guide to creating llms.txt: the exact structure the specification asks for, where to upload the file on WordPress or a static host, and a short checklist to validate it.

If you would rather not write it by hand, our free AI visibility audit drafts an llms.txt for your site automatically, as one of its ready-to-paste fixes.

Frequently asked questions

Does llms.txt help SEO?

No — there is no evidence that llms.txt affects Google rankings, and Google has said it does not use the file. llms.txt speaks to AI assistants, not to search ranking. Your titles, content, internal links, and Schema markup still do the SEO work.

What is llms-full.txt?

llms-full.txt is the bigger sibling from the same specification. Instead of linking out to pages, it contains the full text of your content in one file, so a model can read everything in a single request. Documentation sites use it most, because a coding assistant can load an entire manual at once.

Where do I put the llms.txt file?

The file belongs at the root of your domain — yoursite.com/llms.txt — the same place robots.txt lives. AI tools look for it at that exact address, so a file buried in a subfolder will simply never be found.

Is your site visible to AI at all? Check before you optimize

llms.txt is a small piece of a much bigger question: when a customer asks ChatGPT or Claude for the best option in your market, does the answer name you — or your competitor? That is not something to guess about. It is something to check.

Our free audit — run it here — asks AI assistants real questions about your market, shows you exactly which brands they name, and hands you ready-to-paste fixes, an llms.txt draft included. It takes about three minutes, and what you get is a verifiable fact, not an abstract score. The same question has a manual answer too: how to check if AI mentions your brand walks through the thirty-minute version — ten questions and a spreadsheet.