What the Heck Is llms.txt, and Does Your Website Need One?
An llms.txt file is a short text page, published at yourwebsite.com/llms.txt, that lists your best pages for AI tools. It's a 2024 proposal, not a standard. Google says Search doesn't use it, and I couldn't find an AI company saying it changes what it mentions. For most small businesses, it's optional.
Yep, that's the short version. If an SEO tool flagged a missing llms.txt and now you feel behind, you're not. Here's what the file is, what it can't do, and where I'd spend your time first.
What is an llms.txt file?
It's a text file, written in Markdown (plain text with a few simple symbols for headings and links), that sits at the top level of a website. In effect, it's a short reading list you write by hand. The name comes from LLMs, the large language models behind tools like ChatGPT, Claude and Gemini.
Jeremy Howard, co-founder of Answer.AI, proposed it in September 2024, and the original proposal at llmstxt.org is where the format comes from. The idea is that robots.txt sets access rules, while llms.txt offers "a curated overview for LLMs" that a tool can use on demand. The page itself says "We propose adding a /llms.txt markdown file." It isn't a formal standard, and no search engine has committed to it.
What goes inside an llms.txt file?
Per the proposal, the only required part is a heading with the name of your site or business. Then come a short summary in a quote block, optional notes, and sections of links, each with a one-line note about the page. Here's a made-up example for a hair salon:
# Example Hair Studio
> A hair salon offering cuts, color and bridal styling by appointment.
## Services
* [Services and prices](https://example.com/services): what we offer and what each costs
* [Book an appointment](https://example.com/book): online booking and cancellation policy
robots.txt vs sitemap.xml vs llms.txt: which one does what?
robots.txt sets crawling rules, sitemap.xml lists your pages for search engines, and llms.txt is a reading list for AI tools. Only robots.txt is about access. Here they are side by side.
| File | Where it lives | What it's for | What it can't do |
|---|---|---|---|
| robots.txt | yourwebsite.com/robots.txt | Tells crawlers which parts of your site they may visit | Keep a page out of Google's results, or stop a crawler that ignores it |
| sitemap.xml | Usually yourwebsite.com/sitemap.xml | Lists your pages so search engines can find them | Guarantee a page gets crawled or indexed |
| llms.txt | yourwebsite.com/llms.txt | A short curated reading list for AI tools that choose to read it | Block or allow anything |
Google is clear that robots.txt isn't a way to keep a page out of Google. It's a set of requests that well-behaved crawlers follow, and it isn't security. Google also says a sitemap doesn't guarantee everything gets crawled and indexed, and that a small, well-linked site may not need one.
If you want to allow or block a particular AI crawler, robots.txt is where you set it, and llms.txt can't do it. Companies publish their bot names. OpenAI lists OAI-SearchBot for ChatGPT search and GPTBot for model training, and says a site that opts out of OAI-SearchBot won't be shown in ChatGPT search answers. Anthropic and Perplexity publish theirs too. Bots that fetch a page because a person asked, like ChatGPT-User and Perplexity-User, may not follow robots.txt at all. Allowing a bot only makes a page eligible, and it doesn't promise a mention.
Do ChatGPT, Google or any other AI tool actually use llms.txt?
Google says no for Search, and I couldn't find any AI company saying yes. Here's where things stand as of late September 2026.
- Google Search. Google's guide to generative AI features says you don't need new machine readable files or AI text files to appear in Google Search, "as Google Search itself doesn't use them." A later note adds that these files won't help or hurt your visibility, and that it's fine to keep them for other systems that use them.
- Google's John Mueller. In June 2025 he wrote, "FWIW no AI system currently uses llms.txt." That's one person, more than a year ago, so it's a data point, not the last word.
- OpenAI, Anthropic, Perplexity and Microsoft. I couldn't find an official statement from any of them saying their products read llms.txt. Their crawler docs are about robots.txt.
- Studies. Both of these come from SEO software companies. Ahrefs studied about 137,000 domains (May 2026 data) and found that nearly all llms.txt files got no requests that month. Its sample leans toward technical, SEO-aware sites, and a request in a server log isn't proof of use. SE Ranking's November 2025 study of about 300,000 domains found no relationship between having the file and how often a domain is cited by AI tools.
- The wrinkle. Search Engine Journal reported that Chrome's Lighthouse tool now includes an audit for the file, so a tool may flag yours as missing. It describes that audit as separate from search ranking.
- Where it may get used. Some developer documentation sites publish these files, and Ahrefs noted that coding agents show real interest. A software company with a big docs site is a different case from a salon.
What I know: some sites publish it, and Google says it ignores it. What I don't know: whether some AI tool reads it in some situation. I can't see inside those tools, and I'd be wary of anyone who says they can.
Does my small business website need an llms.txt file?
No. If you're a stylist, a contractor, a coach or a baker and customers find you through Google, your Google Business Profile, Pinterest or referrals, nobody has measured a payoff from an llms.txt file for a business like yours. I'd spend your first free hour on the basics below instead.
If you still want one, go ahead as long as it's quick to write and you'll keep it accurate. If your services, prices or hours change, the file has to change too. A stale llms.txt describing last year's offers is the only real downside I can see. I don't know whether every website builder lets you publish a file at /llms.txt, so ask whoever built your site.
Do I keep an llms.txt on marieglowstudio.com?
I do, and you can read it. It follows the proposal's shape: the studio's name as the heading, a one-sentence summary in a quote block, a short paragraph on how the work gets done, then linked sections with a note on each. Those cover services, who the studio is for, the studio pages, blog topics, comparisons and guides, plus a reference section pointing to my pricing file, a longer companion called llms-full.txt, my Pinterest profile and my Google Business Profile. The blog and comparison lists are regenerated by scripts, so they stay in step with the site.
Why keep it? It's a tidy summary of the site that costs almost nothing to maintain, and Google says keeping one is fine. I'm not counting on it for anything. The files with documented jobs matter more: my robots.txt names a handful of AI crawlers, including GPTBot, OAI-SearchBot, ClaudeBot, Claude-SearchBot and PerplexityBot, allows them, and points to my sitemap.xml, which lists the pages.
What should I do before worrying about llms.txt?
Spend your effort on what you control. I can't promise you a mention in ChatGPT, Perplexity or Google's AI features, and neither can anyone else. Google's own guide says SEO fundamentals still apply, and that's where I'd start.
- Make sure search engines can reach and index your pages. If your site is missing from results entirely, start with why a website is not on Google.
- Write pages that state plainly what you do, where, for whom and at what price, and keep those facts identical across your site, your Google Business Profile and other listings. Clear, accurate information is easier for AI tools to work from, and it helps human visitors too.
- Add structured data where it fits. What schema markup is explains what it does and whether you need it.
- Get the vocabulary straight. SEO, AEO and GEO sorts out the acronyms, and answer engine optimization for small business covers writing pages that are easy to quote.
- Monitor what's happening. Google Search Console offers a generative AI performance report, and Bing Webmaster Tools launched an AI Performance report in public preview in February 2026. They don't promise anything, but they're free and show real data about your site.
If you'd rather have someone look at the basics for you, the Website SEO Audit reviews your technical setup (including your sitemap and robots.txt), on-page SEO, local SEO and AI-search readiness, then hands you a written, prioritized fix list. It costs $247 and takes about 7 business days. It's the review and the plan, and the SEO Foundations Setup is the version where the work gets done for you.
Frequently asked questions
Is llms.txt the same as robots.txt?
No. robots.txt tells crawlers which parts of your site they may visit, and it's the file you use to allow or block a specific bot. llms.txt is an optional, hand-written reading list for AI tools. It can't block or allow anything, and it doesn't replace robots.txt or sitemap.xml.
Can llms.txt block AI crawlers like GPTBot or ClaudeBot?
No. Blocking or allowing a crawler is done in robots.txt, using the bot names each company publishes. An llms.txt file only offers a reading list. Not every fetcher follows robots.txt, though: bots that load a page because a person asked, like ChatGPT-User and Perplexity-User, may not, so check each company's page before you change anything.
Does ChatGPT read llms.txt?
I don't know, and I couldn't find OpenAI saying that it does. OpenAI documents robots.txt controls for OAI-SearchBot and GPTBot, and says robots.txt rules may not apply to ChatGPT-User, which fetches pages when a user asks. None of that mentions llms.txt.
Can an llms.txt file hurt my website?
Google says these files won't help or hurt your visibility in Search, so a correct one is low risk. The one downside I can see is a stale file that still describes old services, prices or hours. If you publish one, update it whenever the facts on your site change.


