Short answer
llms.txt is a short Markdown file at yourstore.com/llms.txt that gives AI tools a map of your site: who you are, and links to your key pages and policies. It is a proposal, not a standard. Google says Google Search does not use it, and we found no OpenAI, Anthropic or Perplexity documentation saying their bots read it, so treat it as a cheap extra, not a ranking lever.
Key takeaways
- llms.txt is a proposal from 2024, not an adopted standard.
- Google Search says it ignores the file, for better or worse.
- A store version takes minutes: name, summary, products, policies.
An llms.txt file is a plain-text summary of your store, written in Markdown, that sits at a fixed address so AI tools can find it. It takes a few minutes to write. Whether any given AI system reads it is a separate question, and the honest answer is "mostly unconfirmed".
This guide covers what the file is, what the big AI companies have said about it, how to write one for a shop, and where to put it.
What is llms.txt?
llms.txt is a proposed file that gives AI tools a short, curated map of a website. It was proposed by Jeremy Howard on llmstxt.org, first dated 3 September 2024 and since revised as "v2".
The proposal's own words are: "We propose adding a /llms.txt markdown file to websites to provide LLM-friendly content." The idea is simple. Web pages are full of menus, scripts and layout. A language model reading your store would rather have a clean summary and a list of the pages that matter.
The file is meant to be read on demand. The proposal says llms.txt information "is instead used on demand, when an agent needs information about a topic while assisting a user." That is different from a search crawler indexing your whole site.
It is not the same as robots.txt. robots.txt tells bots what they may and may not access. llms.txt does not block or allow anything. It only describes. If you want to control access, see robots.txt for AI agents.
Do AI companies actually read llms.txt?
Not that they have said, for the most part. llms.txt is a proposal, not a standard adopted by a standards body. Of the four companies below, only Google has made a clear statement about it, and that statement is a no.
Here is what each company's own documentation says, at the time of writing:
| Company | What its own docs say about reading your llms.txt |
|---|---|
| Google (Search) | Says Google Search, including its generative AI features, does not use it |
| OpenAI | Crawler docs describe GPTBot, OAI-SearchBot and ChatGPT-User with no statement that they read it |
| Anthropic | Crawler docs describe ClaudeBot, Claude-User and Claude-SearchBot and do not mention llms.txt |
| Perplexity | Crawler docs describe PerplexityBot and Perplexity-User with no statement that they read it |
Google is the clearest. Its guide to generative AI features in Search says: "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." Google added a note on 15 June 2026 clarifying this, and the guide also says creating the file "will neither harm nor help your site's visibility or rankings in Google Search, as Google Search ignores them."
There are two points on the other side. First, OpenAI, Anthropic and Perplexity each publish an llms.txt for their own developer documentation. Publishing a file for your own docs is not the same as reading other sites' files, so don't read too much into it. Second, Google's Chrome team includes an llms.txt check in Lighthouse's agentic browsing audits. Lighthouse's page says "providing the file is optional at the moment", and a missing file is marked not applicable rather than failed.
So what's the verdict? llms.txt is a low-cost file that some AI tools and coding assistants may use when a person points them at your site. It is not a known ranking signal for any major AI search product.
What should an online store put in llms.txt?
Put the few things an AI assistant would need to answer a shopper's question correctly: who you are, what you sell, where you ship, and links to your products and policies. Keep it short.
The proposal sets out a fixed order:
- An H1 with your store's name. This is the only required part.
- A blockquote with a short summary. One or two sentences: what you sell and to whom.
- Optional plain paragraphs. Short facts that apply everywhere, such as tax or made-to-order times. No headings here.
- H2 sections with lists of links. Each item is a Markdown link, then optionally a colon and a short note.
- An "Optional" section. By convention, links an AI tool can skip when it needs to be brief.
For a store, good sections are Products (your main collections), Policies (shipping, returns, contact) and maybe Help (sizing, care guides). Use full URLs.
What does a store llms.txt look like?
Here is a made-up example for a small candle shop. Swap in your own name, facts and links.
# Hollow Oak Candles
> Hand-poured soy candles made in the UK. We ship to the UK and EU.
All prices include VAT. Items marked "made to order" ship in 5 to 7 working days.
## Products
- [All candles](https://example.com/collections/all): Full range with prices and scents
- [Gift sets](https://example.com/collections/gift-sets): Boxed sets of three candles
## Policies
- [Shipping](https://example.com/policies/shipping-policy): Rates, delivery times and countries
- [Returns](https://example.com/policies/refund-policy): How to return unused items
- [Contact](https://example.com/pages/contact): Email and response times
## Optional
- [About us](https://example.com/pages/about)
- [Candle care guide](https://example.com/blogs/guides/candle-care)
A few rules of thumb:
- Link to collections, not every product. The proposal describes a curated overview. A list of thousands of products is what a sitemap is for.
- Only state facts that are true on your site. If your returns window changes, the file must change too. A stale llms.txt gives AI tools wrong answers about your shop.
- Skip marketing copy. "The best candles in the world" tells a model nothing. "Soy wax, 40-hour burn time" does.
- Don't put anything private in it. The file is public, like any page.
The proposal also suggests offering clean Markdown copies of key pages, for example page.md next to page.html. That is optional, and may need help from a developer.
Where do you put llms.txt?
Put it at the root of your domain, so it loads at https://yourstore.com/llms.txt. That is the location the proposal names, following the same pattern as /robots.txt.
How you get it there depends on your platform:
- Self-hosted sites and WordPress. Upload the file to your web root, the same folder as robots.txt, or use a plugin that serves it.
- Hosted store platforms. Some don't let you place a file at the root of your domain. Check your platform's help pages, or use an app that serves the file for you.
The proposal also allows a file at a subpath, such as /shop/llms.txt. That file only covers the pages under its own path, so a root file is better for a whole store.
After you publish it, open the address in a browser. You should see plain text, not your store's 404 page. If characters look garbled, save the file again as UTF-8 text. You can also run the free scan at /scan, which checks whether an llms.txt file is present along with other files agents read.
How do you keep llms.txt up to date?
Treat it like your shipping page: review it whenever your policies, collections or shipping countries change. Set a reminder to check it every quarter.
The biggest risk with llms.txt is not a missing file. It is a wrong one. If the file says you ship to Canada and you stopped last month, any tool that trusts it will repeat the mistake to a shopper.
How Sill helps. Sill publishes the files AI agents read, including llms.txt, an agent card, an MCP server and a catalog, and keeps them current. It also shows which AI agents and bots visit your site and which pages they read, so you can see what is being fetched instead of guessing.
Should your store bother with llms.txt?
Yes, if it takes you less than an hour, and no, if you're expecting it to move rankings. Google has said Search ignores it, and no other major AI company has said in its own docs that its bots read it. It costs little, does no harm in Google's own words, and some tools and coding assistants may use it.
Spend most of your effort where the evidence is stronger: complete product data, clear policies, and the right robots.txt rules. Those matter for search engines and shopping agents alike, and they are part of being ready for agentic commerce.
Questions people ask
Will llms.txt help my store show up in ChatGPT or Google AI Overviews?
There is no primary source saying it will. Google says Google Search, including its generative AI features such as AI Overviews and AI Mode, doesn't use files like llms.txt, and OpenAI's crawler documentation does not say its bots read it.
Can an llms.txt file hurt my SEO?
Google says creating one will neither harm nor help your visibility or rankings in Google Search, because Google Search ignores it.
Is llms.txt the same as robots.txt?
No. robots.txt tells bots what they may access. llms.txt is a summary with links, meant to be read when an AI tool needs information about your site. It does not block or allow anything.
Should I list every product in llms.txt?
Usually not. The proposal describes a curated overview, not a full list like a sitemap. Link to your collections and key policies instead.
Sources
- llmstxt.org: The /llms.txt file, v2 · checked 16 Nov 2026
- Google Search Central: Optimizing your website for generative AI features on Google Search · checked 16 Nov 2026
- Google Search Central: Latest Google Search documentation updates · checked 16 Nov 2026
- Chrome for Developers: Lighthouse llms.txt audit · checked 16 Nov 2026
- OpenAI: Overview of OpenAI crawlers · checked 16 Nov 2026
- Anthropic: Does Anthropic crawl the web and how can site owners block the crawler? · checked 16 Nov 2026
- Perplexity: Perplexity crawlers · checked 16 Nov 2026
- OpenAI developer docs llms.txt · checked 16 Nov 2026
- Anthropic developer docs llms.txt · checked 16 Nov 2026
- Perplexity docs llms.txt · checked 16 Nov 2026