What is llms.txt, and does your site need one?
llms.txt is a proposed file that summarizes your site for AI tools. Here's what it is, who actually uses it, and whether it's worth adding.
On this page10 sections
Short answer
llms.txt is a proposed Markdown file at the root of your site that gives AI tools a short summary and a list of your most useful pages. Google says Search doesn't use it, and no major AI search provider has said its crawler relies on it. It's cheap to add and handy for documentation-heavy sites, but it isn't a ranking factor and it won't get you into AI answers by itself.
If you've been to a marketing conference or opened LinkedIn in the past year, someone has told you to add an llms.txt file. Some people call it "robots.txt for AI." It isn't, and the confusion is costing businesses time and money. Here's what the file actually does, who reads it as of August 2026, and how to decide whether your site needs one.
What is llms.txt?
llms.txt is a proposal, not an official web standard. It was published at llmstxt.org by Jeremy Howard in September 2024. The idea is simple: websites are built for people, with navigation, scripts, cookie banners and layout code. Large language models work better with clean text, and they have limited room for context. So a site can offer a single Markdown file at /llms.txt that says, in effect, "here's who we are, and here are the pages worth reading."
The proposal also suggests offering clean Markdown versions of important pages, at the same URL with .md added, so a tool can read the content without the page chrome.
That's it. There's no registration, no verification, and no promise that any AI system will read it.
What goes in an llms.txt file
The format is plain Markdown with a few conventions. Per the spec:
- An H1 with the name of the site or project. This is the only required part.
- A blockquote with a short summary containing the key information.
- Optional paragraphs or lists with more detail.
- H2 sections, each containing a list of links with a short note on what each page covers.
- By convention, a section called "Optional" for secondary links a tool can skip if it's short on space.
Here's what one might look like for a hypothetical plumbing company:
# Northside Plumbing Co.
> Licensed plumbing contractor serving the north suburbs of Chicago since 2009. Residential repair, water heaters, sewer lines and remodels. Emergency service 24/7.
## Services
- [Water heater repair and replacement](https://example.com/water-heaters): tank and tankless, typical price ranges, same-day availability
- [Sewer line repair](https://example.com/sewer): camera inspection, trenchless options, what affects cost
- [Bathroom and kitchen remodels](https://example.com/remodels): process, timelines, permits
## Company
- [Service area](https://example.com/service-area): towns we cover and travel fees
- [About and licensing](https://example.com/about): license numbers, insurance, team
## Optional
- [Blog](https://example.com/blog): maintenance guides and seasonal tipsNotice it's a curated table of contents with context, not a copy of the whole site, and not a set of rules.
Who actually uses llms.txt
This is where the hype and the facts part ways. Here's what we could verify as of early August 2026.
Google Search doesn't use it. Google's guide to optimizing for generative AI features is unusually direct: 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." It adds that creating such files "will neither harm nor help your site's visibility or rankings in Google Search."
OpenAI's publisher guidance doesn't mention it. OpenAI's FAQ for publishers says any public website can appear in ChatGPT search, and the step it asks for is making sure you aren't blocking its search crawler, OAI-SearchBot, in robots.txt. Nothing about llms.txt.
We found no public statement from a major AI search provider saying its crawler reads llms.txt to decide what to cite. If that changes, it'll be big news, and we'll update this post.
AI companies do publish their own. The llmstxt.org site notes that OpenAI, Anthropic and Google's Gemini team publish llms.txt files for their developer documentation. That's the real use case: developer docs that people feed into coding assistants and agents. When a developer points a tool at a docs site's llms.txt, the tool gets a clean map of the documentation instead of scraping a hundred HTML pages.
So the file gets read when someone, or some tool, goes looking for it on purpose. It isn't (as far as anyone has shown) a signal that search systems collect from every site.
llms.txt vs robots.txt vs sitemap.xml
The three files get lumped together. They do different jobs.
| File | What it does | Who reads it | Does it control access? |
|---|---|---|---|
| robots.txt | Tells crawlers which paths they may crawl | Search and AI crawlers that follow the standard | Yes, for crawlers that respect it |
| sitemap.xml | Lists the URLs you want discovered | Search engines | No |
| llms.txt | Summarizes the site and points to key pages | Tools that choose to fetch it | No |
The last column matters most. If you want to stop an AI company's crawler, llms.txt does nothing. That's robots.txt's job, using the specific user agent names each company publishes, such as Google-Extended for some of Google's AI systems or GPTBot for OpenAI's training crawler.
Is llms.txt a ranking factor?
No. For Google, that's Google's own statement. For other systems, there's no evidence it influences what gets cited. What does influence AI answers is the same thing that influences search: pages that can be crawled, that answer real questions specifically, and that come from a source others trust. We cover that in how to get cited in Google's AI Overviews and how to show up in ChatGPT answers.
Should your site have one?
Our honest take, by type of site:
- Software, APIs and documentation sites: Yes. Developers really do use these files with coding tools, and maintaining one alongside your docs is low effort.
- Large catalogs and knowledge bases: Probably worth it, especially if people use AI tools to research your products in detail. Keep it to the pages that matter.
- Small business and local service sites: Optional. It takes under an hour to write, it can't hurt, and it won't move your leads. Do the basics first.
The risk isn't harm from having the file. The risk is spending money on it instead of on the things that work, or letting it go stale so it contradicts your site. A file that lists services you stopped offering is worse than no file.
Why we serve one anyway
Simply Works' own site serves an /llms.txt file. We added it because it's cheap, because it forces a useful discipline (can you summarize what you do in one paragraph and pick your ten most important pages?), and because AI agents that do go looking get an accurate summary instead of guessing. We don't expect it to bring in clients, and we wouldn't sell it to you as if it would.
How to add an llms.txt file
If you decide to add one, this is the process we'd follow:
- Write the summary first. One or two sentences: who you are, what you do, where. Put it in the blockquote.
- Pick the pages. Your main service or product pages, pricing or process pages, service area, about and contact. Ten to thirty links is plenty for most businesses.
- Write a one-line note for each link that says what's on the page in plain words.
- Save it as
llms.txtat your site root so it loads atyourdomain.com/llms.txtas plain text. - Check it renders in a browser as text, not as a download or an error page.
- Add it to your update routine. When you add or retire a service page, update the file.
Most CMSs can serve a static text file, and some tools will generate one for you; Yoast, for example, added an llms.txt generator to its SEO plugin in 2025. A generated file is a fine start, but read it and trim it. If yours can't, your web developer can add it in a few minutes.
Where to start
If you're thinking about llms.txt because you're worried about AI search, start with the things that have evidence behind them: make sure your robots.txt isn't accidentally blocking search crawlers, make sure your key pages explain your services in specific, plain text, and keep your Business Profile current. Then add llms.txt as a small extra if you like.
If you'd like us to check how your site holds up for search and AI answers, that's part of our SEO and local SEO work. If you're rebuilding the site anyway, our web design team can make llms.txt part of the build.
Frequently asked questions
Is llms.txt the same as robots.txt for AI?
No. robots.txt tells crawlers what they may access. llms.txt only describes your site. It can't allow or block anything.
Do I need llms-full.txt too?
Some documentation sites publish a second file with the full text of their docs in one place. That can help developers using coding tools. Most business sites don't need it.
Will adding llms.txt hurt my SEO?
Google says it will neither harm nor help your visibility in Google Search. Just keep it accurate.
