What is llms.txt?
llms.txt is a plain text file you put at the root of your website, so it lives at yourstore.com/llms.txt. The idea comes from a public proposal at llmstxt.org. It is meant to help AI tools find the pages that matter most, without wading through menus, pop-ups and code.
Think of it like a printed map at the front door of a shop. It does not change the shop. It just points visitors to the right aisles faster.
What goes in an llms.txt file?
The llmstxt.org format uses Markdown, a simple way to write headings and links with plain symbols. The file is short and easy for a person to read too.
A store's file usually has a few clear parts, in this order:
- A title line with your store's name, starting with a # sign.
- A one or two sentence summary of what you sell and who it is for.
- Headings for groups of pages, such as Products, Buying guides or Shipping and returns.
- Under each heading, a list of links, each with a short note on what the page covers.
- An optional section for pages that are nice to have but not essential.
How is llms.txt different from robots.txt?
robots.txt is a different file. It tells bots which parts of your site they may or may not read. If it blocks the wrong bot, that can hide your store from AI answers.
llms.txt does none of that. It cannot let a bot in or keep one out. It only suggests which pages are worth reading. So a good llms.txt will not fix a robots.txt that blocks ChatGPT's search bot.
Do AI assistants actually use llms.txt?
The honest answer is that nobody has shown it makes a clear difference. We have not found a public statement from OpenAI, Google or Perplexity saying their assistants read llms.txt when they answer shoppers. We also have no figure for how many AI tools use it.
The one real-world report we have points the same way. Website builder Webflow saw no significant lift from an llms.txt file and focuses on content instead, according to Growth Unhinged in November 2025. That is one company, and it sells software, not products. But it is a useful reality check.
Some companies do keep an llms.txt alongside plain-text versions of their help pages, as one part of a bigger effort. It is cheap to do. Just do not expect it to move results on its own. There is no secret markup that makes AI recommend you.
Does a Shopify store need an llms.txt?
You do not need one to be recommended. AI assistants still read your normal pages and the wider web. What tends to matter more is clear product pages, other sites mentioning your brand, and bots being free to read your store.
That said, a tidy llms.txt does no harm. It takes little time, and it pushes you to decide which pages best explain your products. That thinking is useful on its own. If you add one, keep it accurate and update it when your range changes.
Before you write one, check whether your store already has a file. Visit yourstore.com/llms.txt in your browser. If text appears, you have one. Read it and see if it lists the right pages.
How to check your llms.txt with a free tool
FirstPicked's free llms.txt Checker reads your file and compares it with the llmstxt.org format. If you do not have a file yet, it gives you a starter file to copy.
Here is a quick way to use it:
- Enter your store's web address in the llms.txt Checker.
- See whether a file was found and whether it follows the format.
- Fix anything flagged, such as a missing title or summary.
- Check every link points to a real, live page on your store.
- If you have no file, copy the starter and edit it with your own pages and plain descriptions.
- Recheck after big changes, like new product ranges or removed pages.
What to fix before you worry about llms.txt
Some jobs have a much bigger effect on whether AI assistants can name your store. Do these first, then add llms.txt as a finishing touch.
Make sure your robots.txt does not block the bots that matter. Googlebot covers Google, including AI Overviews. OAI-SearchBot decides whether you show in ChatGPT search. PerplexityBot does the same for Perplexity. Firewall settings, such as Cloudflare's AI bot blocking, can also block bots and do not show in robots.txt.
- Put who the product is for and its key facts in the first lines of each description.
- Use exact names: materials, ingredients, sizes and who it suits.
- Add question and answer sections that each make sense on their own.
- Keep prices, stock and facts up to date.
- Earn honest mentions on sites AI already reads. Never use fake reviews or bought links: they break the platforms' rules and hurt your store.
Key takeaways
- llms.txt is a short file at yourstore.com/llms.txt that lists your key pages for AI tools.
- It cannot block or allow bots: that is robots.txt's job.
- There is no public proof yet that llms.txt helps a store get recommended, and Webflow reported no significant lift from one.
- It is cheap and harmless, so add a tidy one after you fix robots.txt, product pages and outside mentions.
- A free checker can test your file against the llmstxt.org format and give you a starter to copy.
Questions people ask
Will an llms.txt file get my store recommended by ChatGPT?
Not on its own. There is no public evidence that llms.txt makes AI assistants recommend a store. Webflow reported no significant lift from one in 2025. Clear product pages and honest mentions on other sites tend to matter far more.
Can llms.txt hide my store from AI?
No. llms.txt only suggests pages to read and cannot block anything. If you are worried about being hidden, check your robots.txt and any firewall bot settings instead.
How do I know if my Shopify store already has an llms.txt?
Type yourstore.com/llms.txt into your browser. If text shows up, a file exists. Read it to make sure it lists your real, current pages with accurate descriptions.
What should I put in my store's llms.txt?
Start with your store name and a one or two sentence summary of what you sell. Then add headings like Products and Buying guides, each with links and a short note on what the page covers. Keep it short, accurate and up to date.
Sources
Every number in this article comes from one of these pages. Written by the FirstPicked team with help from AI, and checked against our facts sheet and these sources before publishing. Nothing here is a promise of results.