All Resources
5 min read

llms.txt: The One-File Signal That Tells AI Who You Are

A single llms.txt file gives AI crawlers a clean, authoritative summary of your business. Here's the exact format and why it's the fastest AEO win.

What llms.txt is

llms.txt is a plain text file placed at the root of your domain — yourbusiness.com/llms.txt — that gives AI crawlers a concise, structured summary of your business: who you are, what you do, where you operate, and which pages matter most. The idea is to hand a model the answer directly rather than making it reconstruct your identity from scattered pages, images and scripts. It is a proposal rather than an official standard, and not every provider reads it, but the cost of publishing it is a few minutes and the cost of omitting it is that you rely entirely on inference.

Why it is a fast win, and what it is not

Because it is a single static file, it can be written and crawled within days, which makes it one of the quickest things to ship in an AEO programme. What it cannot do is substitute for the rest. llms.txt does not replace crawlable pages, accurate schema, or independent proof — a model that reads your summary and then finds nothing corroborating it has simply learned that your self-description is unsupported. Treat it as a précis that points at evidence, not as the evidence.

What belongs in the file

Open with the business name and a single sentence saying what it does and for whom. Then your core services, one line each. Then your service area, stated the same way it appears on your site. Then anything that genuinely differentiates you — specialisms, credentials, years operating. Then a short list of your most useful pages with a one-line description of each. Keep it factual and free of marketing adjectives: the file is read as a statement of fact, and adjectives reduce its usefulness because a model cannot verify them.

A worked example

This is the shape to aim for, with placeholders where your facts go. # Your Business Name > One sentence: what you do, for whom, and where. ## Services - Service one: one line of plain description. - Service two: one line of plain description. ## Service area Las Vegas, Nevada, and surrounding areas. We travel to customers rather than serving them at a premises. ## About Founded YEAR. Owner: Name. Specialisms: X and Y. ## Key pages - /services/: what is included and how pricing works - /about/: who does the work and their background - /contact/: how to reach us and how quickly we reply

Keep it consistent with everything else

The facts in llms.txt must match your schema, your visible copy and your directory listings exactly. If the file claims a service area that your site contradicts, you have manufactured a contradiction rather than resolved one — which is worse than having no file at all. This is the same consistency rule that governs citations and structured data, and it is the reason we generate the file from verified facts rather than writing it fresh each time.

Where to put it and how to serve it

The file goes at the domain root: /llms.txt. It must be publicly reachable without authentication and should return a 200 status. Serve it as text/plain so it renders in a browser rather than downloading. Do not put it behind a redirect chain, and do not add it to robots.txt as a disallowed path — that would defeat the purpose. If you also publish an expanded version listing every page, name that file llms-full.txt and keep the root file short, because the concise version is the one most likely to be read.

What we see in practice

The businesses that benefit most are those whose identity is otherwise hard to reconstruct — multi-service companies, businesses with several locations, and service-area businesses with no storefront, where a model has to infer a great deal from prose. Sites that are already simple and well-structured see less change, because the model was already reaching the right conclusion. That is worth knowing before you judge the result: no movement after publishing llms.txt may mean your site was already clear, not that the file failed.

How to check it is working

Fetch yourbusiness.com/llms.txt in a browser and confirm it displays as plain text rather than downloading or erroring. Confirm it contains no placeholder text. Then re-run the same customer prompts you measured before, and look at whether the sources cited in the answers include pages you listed in the file. Provider support varies, so judge it over weeks rather than days, and treat it as one contributing signal rather than something you can attribute a ranking change to on its own.

Don't want to DIY?

We publish and maintain llms.txt for every Click Elevate client as part of standard setup, generated from the same verified business facts used for your schema, so the two can never drift apart.

This guide resolves these report findings

  • llms.txt published
  • llms.txt content quality

Run the free AEO score to see which of these apply to your site.

Did this guide answer your question?

Free Instant Analysis

See how visible your business is to AI

Get your free 0–100 AEO score across ChatGPT, Perplexity, Gemini, Claude, and Google AI Overviews.