What is llms.txt, and does your store need one?
Updated llms.txt is a proposed convention: a short Markdown file at yourstore.com/llms.txt that tells AI assistants what your site is and where the important pages are. Support among AI companies is still limited and uneven, so treat it as a cheap, low-risk addition rather than a way to rank in AI answers.
What llms.txt is
AI assistants such as ChatGPT, Claude and Perplexity increasingly look things up on the web while answering. Store pages are full of menus, scripts, pop-ups and repeated product grids, which makes it harder for a program to pick out what matters.
The llms.txt proposal, described at llmstxt.org, suggests a simple fix: publish one plain file at the root of your site that explains who you are in a few lines and links to your most useful pages. It is written in Markdown, a plain-text format where # marks a heading and - starts a list item, so it is easy for both people and programs to read.
What one looks like
The format is a title, a one-line summary in a quote block, then sections of links with a short note on each:
# Example Linen Co.
> Small-batch linen bedding and throws, made in Portugal and shipped from the UK.
We sell direct to customers in the UK and EU. All products are stonewashed linen.
## Shop
- [Bedding](https://example.com/collections/bedding): duvet covers, sheets and pillowcases
- [Throws](https://example.com/collections/throws): lightweight linen throws
## Help
- [Shipping and returns](https://example.com/policies/shipping-policy): delivery times and 30-day returns
- [Care guide](https://example.com/pages/care): how to wash and dry linen
- [Contact](https://example.com/pages/contact): email and response timesKeep it short and factual. It should match what your site says, because an assistant may compare the two.
How much it helps today
Be realistic about this. llms.txt is a community proposal, not a standard that AI companies have agreed to follow, and support for reading it is still limited and uneven. Publishing one does not guarantee that any assistant will read it, mention your store or rank you higher.
The case for it is that it costs very little: one small file, no effect on your human visitors and no known downside. If support grows, you are ready. What matters more right now is whether AI crawlers are allowed to reach your pages at all, and that is decided by robots.txt.
On Shopify you can't simply upload a file to the root of your domain. Depending on your setup, options include a theme template, a URL redirect or an app. Check that yourstore.com/llms.txt actually loads as plain text once it is in place.
robots.txt and AI crawlers
robots.txt is a long-standing file at the root of your site that tells automated visitors which parts they may fetch. Most AI companies publish the names their crawlers use, so you can allow or block each one. They fall into two broad groups:
- Search and answer crawlers fetch pages so an assistant can find and cite them in answers. Examples include OAI-SearchBot (ChatGPT search) and PerplexityBot (Perplexity).
- Training crawlers collect pages to help train future AI models. Examples include GPTBot (OpenAI), CCBot (Common Crawl, whose data is used by many AI projects) and Google-Extended, which is a setting Google reads rather than a separate crawler. Blocking Google-Extended does not remove you from Google Search.
A robots.txt that welcomes AI search but opts out of training might include:
User-agent: OAI-SearchBot
Allow: /
User-agent: PerplexityBot
Allow: /
User-agent: GPTBot
Disallow: /
User-agent: Google-Extended
Disallow: /
User-agent: CCBot
Disallow: /Choosing your trade-off
Blocking search crawlers keeps your pages out of those assistants' answers, which means shoppers asking for a recommendation won't see your store there. Blocking training crawlers has no effect on being found today, but it is a choice about whether your content helps build future models. Many stores allow search crawlers and decide on training separately.
Whatever you choose, check that your robots.txt says what you think it says. Rules added by apps or old templates sometimes block more than intended. robots.txt is a request that well-behaved crawlers follow, not a lock, so it won't stop a crawler that ignores it.
On Shopify, robots.txt is generated for you and can be customized through a robots.txt template in your theme. The full list of checks shows what else affects how assistants and search engines see your store.