What Is llms.txt (and Should Your Site Have One)?

llms.txt is a plain-text file you place at the root of your website that points AI tools - the large language models behind ChatGPT, Perplexity, and similar engines - to your most important, cleanly-written content. Think of it as a short, curated guide for AI crawlers: instead of leaving them to wade through your entire site, you hand them a tidy map of what matters and where the clean version lives. It is a young convention, not an official standard, but it is simple, free, and increasingly worth having.

There is a lot of confusion about what this file does and does not do, so let me clear it up. In this guide I will explain what llms.txt is, what to put in it, exactly how it differs from robots.txt and sitemap.xml, and whether your small business site actually needs one yet.

What is llms.txt, exactly?

The idea behind llms.txt is that AI engines work best with clean, structured, easy-to-read content, and they often struggle with the clutter of a normal web page: navigation menus, pop-ups, ads, scripts, and styling all get in the way of the actual information. llms.txt is a proposed convention to fix that. It is a Markdown file at yoursite.com/llms.txt that lists your key pages, with short descriptions and links, so an AI crawler can quickly find the content you most want it to understand and cite.

The format is deliberately simple. At minimum it has a title (your business name), a short description of what you do, and a curated list of links to your most important pages, each with a one-line note on what it covers. Some sites also publish a companion llms-full.txt that contains the full clean text of those pages in one place, so an AI does not even have to fetch and strip each page itself.

What to include in an llms.txt

The goal is curation, not completeness. You are choosing the content you most want an AI to draw on, not dumping your whole site. A good llms.txt for a small business typically includes:

• Your business name and a one-line summary of what you do and who you serve.
• Your core service or product pages, each with a short description.
• Your best explanatory content - the guides, FAQs, and answer-first articles that make you a useful source.
• Key reference pages like pricing, process, or about, if they contain facts you want quoted correctly.
• Contact or booking links, so an AI pointing someone to you knows how they can reach you.

What you leave out matters just as much. Skip thin pages, duplicate content, and anything you would not want an AI to treat as a definitive source. The whole value is in pointing engines at your best, clearest material - the same answer-first, well-structured content that powers generative engine optimization.

llms.txt vs robots.txt vs sitemap.xml

These three files all live at your site root and all talk to automated visitors, but they do completely different jobs. Confusing them is the most common mistake, so here is the clean comparison.

The key differences: robots.txt is about permission (what can be accessed), sitemap.xml is about coverage (here is everything, find it all), and llms.txt is about curation (here is what matters most, in clean form). robots.txt and sitemap.xml are long-established and understood by every major search engine. llms.txt is newer and not yet universally honored. You want all three for different reasons - they do not replace one another.

Does llms.txt actually work yet?

Here is the honest answer, because there is hype on both sides. llms.txt is a proposed convention, not an official standard that every AI engine has committed to reading. Adoption is growing and several tools and crawlers are starting to respect it, but you should not expect it to single-handedly get you cited overnight. It is a low-cost, low-risk signal, not a magic switch.

The realistic framing: llms.txt is cheap to add and does no harm, and as AI-driven discovery keeps growing, having a clean, curated map of your best content ready is a sensible bet. But it is one piece of a bigger picture. On its own it will not make a thin or poorly-structured site citable. It works best as the finishing touch on content that is already clear, factual, and answer-first.

Should your site have one?

For most small business sites, my honest recommendation is yes - add one, but treat it as a finishing step, not a starting point. Here is the order that makes sense.

1. First, get the fundamentals right. A fast, crawlable, well-structured site with clear, answer-first content. Without this, llms.txt has nothing good to point at. Start with SEO for small business websites.
2. Then make your best content genuinely citable. Concrete, factual, clearly-structured pages with the right schema, as I describe in how to get your business cited by AI.
3. Then add a focused llms.txt that points AI crawlers at exactly that content.

Done in that order, llms.txt is a smart, low-effort addition. Done first, on a weak site, it is decoration. The file is only as valuable as the content it points to.

The bottom line

llms.txt is a simple, free, curated map that tells AI engines where your best content is and what it covers. It is not a standard yet, it is not a magic shortcut, and it will not rescue a thin site - but it is cheap, harmless, and increasingly sensible to have as more people find businesses through AI. Add it once your content is genuinely worth pointing at, and it becomes a quiet edge in how visible you are to AI tools.

I build sites that are set up to be found and cited by AI from the ground up - clean structure, the right schema, answer-first content, and a focused llms.txt pointing crawlers at your strongest pages. If you want to know whether your site is ready for AI-driven discovery and what would help most, book a call and send me your URL, or reach me through the contact form.