Do You Need an llms.txt File?
No, you do not need an llms.txt file. Google’s own guide says Search ignores the file, and Ahrefs found 97% of them got zero requests in a full month. If you run a Shopify store, you already have one anyway, since Shopify now builds it for you. The only real mistake is paying someone to make it.
Key Points
- Google’s AI optimization guide says llms.txt files “neither harm nor help” your visibility in Search or its AI features.
- Ahrefs checked 137,210 sites in May 2026, and 97% of the llms.txt files on them were never requested by anything.
- Only 1.1% of the requests that did arrive came from AI retrieval bots like OAI-SearchBot and PerplexityBot.
- Every Shopify store has served a default /llms.txt since spring 2026, so there is nothing to build.
- The work that gets a store cited is open crawling, direct answers on the page, and brand mentions.
What Is an llms.txt File?
An llms.txt file is a short markdown file at yourdomain.com/llms.txt that hands AI models a clean summary of your site. Jeremy Howard proposed the llms.txt standard in September 2024. His pitch was that web pages are messy for machines, so you give the model a tidy text version instead.
That pitch makes sense on paper. A model can only hold so much text at once, and the proposal says context windows “are still too small for most websites in their entirety.” A curated file would save the model from digging through your theme code, popups, and cookie banners to find your product copy.
So think of the file as a menu for robots. It lists your store name, one line on what you sell, and links to the pages you most want read.
But a menu only works if somebody picks it up. People also confuse the file with robots.txt, and the two do very different jobs:
| File | What It Does | Who Reads It | Do You Need It? |
|---|---|---|---|
| robots.txt | Tells bots which pages they can crawl | Googlebot, OAI-SearchBot, nearly every bot | Yes |
| sitemap.xml | Lists every page you want indexed | Google, Bing | Yes |
| llms.txt | Offers a summary of your site in markdown | Mostly audit tools and coding agents | No |
The file that deserves your attention is robots.txt, since an llms.txt file cannot block or allow anything. One bad line in robots.txt pulls your store out of AI answers completely, and I cover that in my post on whether you should block AI crawlers.
Do AI Systems Actually Read llms.txt?
AI systems read llms.txt on about 3% of the sites that have one. Ahrefs studied 137,210 domains in May 2026 and found roughly 38,000 of them had a valid file. On 97% of those sites, nothing requested the file all month.
Even on the 3% that did get a visit, the visitor was rarely an AI search engine. Ahrefs wrote that “if you publish an llms.txt file today, the most likely outcome by far is that nothing ever fetches it.”
The same study looked at sites with no file at all, to see if AI bots came knocking anyway. They did not. “Zero requests came from AI bots for llms.txt files that don’t exist,” Ahrefs wrote. “They never go looking.”
Also, keep the adoption number in context. About 28% of the sites in the study had a file, and Ahrefs calls that an upper bound, since its customers follow SEO news more closely than most site owners do.
Who Requests the File
SEO audit tools request llms.txt more than any AI system does. They made up 21.7% of all requests in the Ahrefs data, and about half of those came from Ahrefs’ own crawlers. The tools checking whether you have the file outnumber the AI systems using it.
| Who Asked for llms.txt | Share of Requests |
|---|---|
| SEO audit tools | 21.7% |
| AI coding agents | 10.5% |
| AI training crawlers (GPTBot, ClaudeBot) | 5.3% |
| AI assistants (ChatGPT-User, Claude-User) | 2.5% |
| AI retrieval bots (OAI-SearchBot, PerplexityBot) | 1.1% |
That last row is the one a store owner should care about. Retrieval bots build the index that ChatGPT search and Perplexity answer from, and they sent 233 requests across the whole study. Where that index comes from is its own question, and I answer it in whether ChatGPT uses Bing. If a bot never asks for the file, the file cannot change what that bot says about your store.
The One Group That Uses It
Coding agents are the one group that uses llms.txt on purpose. When a developer asks an AI assistant to write code against an API, the assistant sometimes loads the docs site’s llms.txt to find the right page. That is why OpenAI, Anthropic, and Perplexity each publish one for their developer docs.
Your store is not a docs site, though. No shopper is asking an AI to write code against your product pages, so the one real use case does not apply to you.
You can also run this check on your own site. If your store sits behind any tool that logs requests by user agent, search the last 30 days for llms.txt and see who asked. ChatGPT finds pages through search indexes instead, which I break down in how ChatGPT picks its sources.
Google Says It Ignores llms.txt
Google ignores llms.txt, and it now says so in its own documentation. Google’s guide to optimizing for generative AI says creating the files “will neither harm nor help your site’s visibility or rankings in Google Search, as Google Search ignores them.” The guide lists llms.txt under things you do not need to do.
Before that page existed, the only word from Google came through John Mueller. He called llms.txt “purely speculative for now” and pointed out that the file “has existed for years, yet none of the AI systems use it.”
He also gave the best rule I have seen for when to bother. “When an AI platform that brings you clients complains that it needs the file for your site, then I’d recommend taking the time to create one.” None has asked yet.
One more thing muddies the water. Chrome’s Lighthouse tool added an llms.txt check in 2026, and some people read that as Google changing its mind. But the Lighthouse documentation says a missing file is marked not applicable, “as providing the file is optional at the moment.”
So nothing about the file affects how you show up in Google’s AI answers. The things that do are regular search work, and I list them in my guide to ranking in AI Overviews.
Your Shopify Store Already Has One
Your Shopify store already has an llms.txt file, since Shopify serves one at /llms.txt on every store by default. Shopify’s May 2026 changelog says each store includes a default agents.md file and that “the paths /llms.txt and /llms-full.txt also point to this content by default.”
Most people picture a content menu, and the Shopify version is something else. It is a set of instructions for shopping agents, the AI tools that browse and buy for a person. It tells them where your product and collection data lives, what your store policies are, and how to check out the right way.
That fits where Shopify is headed. The company is building for AI agents that shop on someone’s behalf, which I explain in what agentic commerce is.
Checking yours takes about a minute:
- Open a browser and go to yourstore.com/llms.txt.
- Confirm a text page loads with a heading like “Agent Instructions.”
- Close the tab and get back to your product pages.
You can replace the default with your own template in the theme code editor if you want to. Shopify’s own developer docs advise against it for most stores and say the managed file is all you need. I agree with them.
Just know the default lives on your main domain only. If you sell in several countries through Shopify Markets, there is no separate version per market, and you do not need one.
This also changes the math on effort. The old advice was that an llms.txt file takes 20 minutes and cannot hurt. On Shopify it now takes zero minutes, since the file showed up on its own.
Does llms.txt Help You Get Cited?
No, llms.txt does not help you get cited by ChatGPT, Perplexity, or Google’s AI answers. Ahrefs put it bluntly after reading a month of logs: “If your goal is showing up in ChatGPT, Perplexity, or AI Overviews, an llms.txt file is largely decoration.”
The reason is simple once you see how these tools find pages. They run a search, pull the top results, and read those pages directly. Nothing in that process stops to look for a summary file first.
Even so, the file keeps spreading, and plugins are why. Common Crawl analyzed 584,107 llms.txt files from its July 2026 crawl and found two thirds were produced by a plugin. Wix alone accounted for 41% of them.
That is adoption without usage. Site builders and SEO plugins switched the file on for millions of sites, so the count looks like momentum. The request logs show almost nobody on the other end.
The steps that do move citations are the ones in my guide to ranking in ChatGPT, and none of them involve a new file.
Never Pay Anyone to Optimize It
You should never pay anyone to optimize an llms.txt file. No shopper-facing AI system reads it, and on Shopify the file already exists at no cost. It still shows up in a lot of the AEO and GEO service packages I review, usually priced like it does something.
The problem is how it looks on the proposal. Line one is llms.txt creation and optimization, line two is AI readiness, line three is entity work, and each one carries a price.
So ask one question before you sign. Which AI bot requested that file last month? If the agency cannot show you a log, you have your answer.
A paid llms.txt line item is a bill for something Shopify already did. That is why the file is not part of my AI search services, and why I tell clients to cross it off other proposals too.
Still, plenty of smart SEOs disagree with me and treat early adoption as cheap insurance. That is a fair view. Cheap insurance should cost you nothing, and for a Shopify store it now does.
What Gets a Store Cited Instead
Three things get a store cited instead of an llms.txt file: open crawling, direct answers on the page, and brand mentions across the web. All three live on pages people can visit. None of them live in a file you upload once and forget.
Start with these, in this order:
- Access: open yourstore.com/robots.txt and make sure nobody blocked OAI-SearchBot, PerplexityBot, or Googlebot. A blocked bot cannot cite you.
- Structure: put the direct answer in the first lines under each heading. Models read your page like an outline and quote the part that answers the question.
- Mentions: earn your brand name in gift guides, review sites, Reddit threads, and press. Models learn who you are from what the rest of the web says about you.
Mentions are the slow one, and they matter most. For a store they come from a gift guide that lists your product, a forum thread where a customer names you, and a local paper that covers your launch. I compare the two signals in brand mentions vs backlinks.
Spend your AI search hours earning mentions, not formatting text files. A store with ten strong mentions and no llms.txt beats a store with a perfect file and no mentions every time.
From there, give it a quarter before you judge. Ask ChatGPT and Perplexity the five questions your best customers ask before they buy, and write down which stores get named. Run the same five questions 90 days later. That list tells you more about your AI visibility than any file on your server.
Four Quick Questions About llms.txt
Do You Also Need llms-full.txt or agents.md?
No, you do not need llms-full.txt or agents.md either. On Shopify all three paths are served by default from the same managed content. On other platforms the same logic applies as with llms.txt, since the bots that send shoppers do not request them.
Does a Missing llms.txt Hurt Your Lighthouse Score?
No, a missing llms.txt does not hurt your Lighthouse score. Chrome’s documentation says the check is marked not applicable when the file is missing. Your speed scores and your rankings are not touched by it.
Can an llms.txt File Hurt Your SEO?
No, an llms.txt file cannot hurt your SEO, since Google says the file will neither harm nor help. The only real cost is time or money spent on it. If a plugin already made one for you, leave it alone.
Will llms.txt Matter Later?
llms.txt might matter later for AI agents, but nothing today says it will matter for search. The proposal was rewritten in 2026 around helping agents use a website, which is the job Shopify’s default file already does. Follow Mueller’s rule and wait until a platform that sends you buyers asks for the file.
Want Your Store Cited in AI Answers?
You do not need an llms.txt file, and on Shopify you already have one. Put your effort into crawlable pages, direct answers, and brand mentions, since that is what the logs and Google’s own guide point to.
If you want a second set of eyes, book a call and I will tell you which of those three is holding your store back and what I would fix first. If you want the ongoing work handled, that is what my AI search services are for.
Sources
- Google Search Central: Optimizing for generative AI features on Google Search
- Ahrefs: We analyzed 137K sites, 97% of llms.txt files never get read
- llmstxt.org: The /llms.txt file proposal
- Shopify changelog: Customize /llms.txt, /llms-full.txt and /agents.md
- Common Crawl: A content analysis of llms.txt files from the July 2026 crawl archive
- Search Engine Journal: Google says llms.txt is purely speculative for now
- Chrome for Developers: Lighthouse llms.txt audit
