llms.txt for Shopify:
What It Does, What Reads It and What to Do Instead

llms.txt for Shopify is a plain markdown file at yourstore.com/llms.txt that summarizes your store for AI tools. As of October 2026, Shopify serves one for every store by default, mirroring a Shopify-generated /agents.md file. You can customize it with a theme template. Evidence that ChatGPT, Perplexity or Google read it is thin, so spend most of your effort elsewhere.

This post covers what the file is, what the data says about who reads it, how Shopify handles it now, and the robots.txt rules that have more effect on AI visibility. It sits inside our full AI search guide for Shopify brands.

What is llms.txt?

llms.txt is a proposal, not a standard. Jeremy Howard published it on llmstxt.org on September 3, 2024. The idea: websites are too big and messy for a language model's context window, so give models one short markdown file with the key facts and links.

The format is simple:

• An H1 with the site name. This is the only required part.
• A blockquote with a short summary.
• Optional paragraphs or lists with more detail.
• H2 sections, each a list of links in [name](url): notes form.
• An optional H2 called "Optional" for links a model can skip when space is short.

The proposal says the file is mainly meant for inference, when a tool fetches it to answer a question, and that future training runs "could" use it too. A companion file, llms-full.txt, holds the full content in one document.

Nothing in the proposal obliges any AI company to read the file. That is the part most "AI SEO" sales pitches skip.

Which AI crawlers actually read llms.txt?

Very few, based on public data. Here is what is documented and what has been observed, as of October 2026.

OpenAI and Perplexity do publish llms.txt files for their own developer docs. That shows they find the format useful for documentation. It does not show their crawlers read yours.

The Ahrefs study, published June 15, 2026, is the best public data. Of the files that did get requests, AI bots made 19.5% of them. GPTBot, OpenAI's training crawler, made 4.51%. OAI-SearchBot, the crawler behind ChatGPT search, made 0.74%. SEO audit tools made more requests than all AI bots combined. Ahrefs notes every figure is a ceiling, because a fetch does not mean the file was used.

Does llms.txt help you show up in ChatGPT or Google AI Mode?

No one has shown that it does. Google says outright that it does not affect Google Search, which includes AI Overviews and AI Mode. Its AI features page adds: "You don't need to create new machine readable files, AI text files, or markup to appear in these features."

For ChatGPT, OpenAI's documented lever is OAI-SearchBot access in robots.txt. Sites that block it "will not be shown in ChatGPT search answers," per its crawler page. Shopify products also reach ChatGPT through Shopify Catalog, covered in our post on getting products into ChatGPT shopping.

Where llms.txt may help is narrower. AI agents and coding tools sometimes fetch it when a user points them at a site. The Ahrefs data shows Claude Code fetching more llms.txt files than most AI retrieval bots. That is a developer use case, not a shopper one.

Our view: a clean llms.txt costs 30 minutes and does no harm. Treat it as hygiene. Do not pay a retainer for it.

How does llms.txt work on Shopify now?

Shopify changed this in 2026. Per the Shopify changelog of May 28, 2026, every store now serves a default /agents.md. By default, /llms.txt and /llms-full.txt serve the same content.

Shopify's agents.md template docs describe agents.md as the store's agent-facing description. It points agents to commerce endpoints, browsing URLs and published policies. Shopify says the default is all most stores need.

The serving order for /llms.txt, per the llms.txt.liquid docs:

1. templates/llms.txt.liquid, if your theme has one
2. templates/agents.md.liquid, if your theme has one
3. Shopify's generated default

This matters if you set up llms.txt before May 2026. Many stores used a URL redirect, a proxy or an app to serve their own file. Merchants in the Shopify developer community reported those methods stopped working once Shopify took over the routes. Check yours today: open yourstore.com/llms.txt and yourstore.com/agents.md in a browser.

How do you customize llms.txt on a Shopify store?

Only do this if you have something the default lacks, such as a short brand summary and links to your best guides. Steps:

1. In Shopify admin, go to Online Store > Themes. Duplicate your live theme first.
2. On the live theme, click the three dots, then Edit code.
3. Under Templates, add a new template. Name it llms.txt.liquid to change only /llms.txt, or agents.md.liquid to change all three files.
4. Write plain markdown. It cannot be a JSON template.
5. Save, then load yourstore.com/llms.txt to check the output.

Two limits to know. The template runs with a restricted Liquid context. Only the request and agents objects exist, so shop, collections and product loops render blank. You cannot auto-list your catalog. Second, Shopify's sample template keeps the commerce discovery, MCP endpoint and sitemap lines from the agents object. Keep them.

A sensible llms.txt for a fictional $1.5M skincare brand:

# Northfield Skin

> US skincare brand selling fragrance-free moisturizers and SPF for sensitive skin. Ships to all 50 states. 60-day returns.

## Key pages
- [Shop all](https://example.com/collections/all): full catalog
- [Ingredient glossary](https://example.com/pages/ingredients): every ingredient and why it is used
- [Shipping policy](https://example.com/policies/shipping-policy)
- [Refund policy](https://example.com/policies/refund-policy)

Do not paste your whole catalog. Do not add claims you would not put on the PDP.

What robots.txt rules should a Shopify store set for AI crawlers?

This matters more than llms.txt. robots.txt is the control every major AI company documents. Each owner splits its crawlers by purpose, so you can allow search and block training separately.

Sources: OpenAI, Perplexity, Google (updated October 15, 2026), Anthropic.

For most D2C brands the right default is to allow all the search and user agents. Whether to block training crawlers is a business call. Blocking GPTBot does not remove you from ChatGPT search, per OpenAI. OpenAI and Perplexity both say changes can take about 24 hours to apply.

How do you edit robots.txt on Shopify?

Shopify generates robots.txt for you. To change it, follow Shopify's robots.txt.liquid guide:

1. In Edit code, add a new template named robots.txt.liquid. This replaces the generated file.
2. Keep the default Liquid loop that prints Shopify's standard rules.
3. Add new user-agent groups outside that loop, for example User-agent: GPTBot then Disallow: /.
4. Before publishing, save your current /robots.txt output and compare it with the new one.

Shopify warns that crawl rule changes can affect indexing. One wrong Disallow: / under User-agent: * blocks Googlebot too. Check what your store serves now at yourstore.com/robots.txt, and look for each AI user agent by name. Also check any bot protection in front of your store. Google's AI features page lists "ensuring that crawling is allowed in robots.txt, and by any CDN or hosting infrastructure" as a best practice.

What should you do instead of obsessing over llms.txt?

Put the hours where platforms have documented an effect:

• Crawl access. Allow OAI-SearchBot, PerplexityBot, Claude-SearchBot, Googlebot and Bingbot.
• Product data. Complete titles, descriptions and category attributes in Shopify Catalog. See our post on product descriptions for AI search.
• Structured data. Accurate price, stock, ratings and policies. See Shopify product schema.
• Feeds. Merchant Center for Google. See Google AI Mode shopping and our Google Shopping guide for D2C.
• Measurement. Track AI referrals so you know if any of this works. See track ChatGPT traffic in GA4 and our GA4 setup for D2C.

sorted.agency publishes its own llms.txt and gets visits referred from chatgpt.com. We cannot attribute those visits to the file, and we would not claim to. If you want a second pair of eyes on crawl access and product data, book a call.