What Is llms.txt? A Shopify Merchant's Guide

llms.txt is a plain-Markdown file you place at your domain’s root — yourstore.com/llms.txt — that gives AI models a curated map of your most important pages. Proposed in September 2024, it is meant to help assistants like ChatGPT, Claude, and Perplexity understand your store. As of mid-2026, though, no major AI company has confirmed it actually reads the file.

That gap is exactly why this guide exists. AI is now a real shopping channel, not an experiment: in Salesforce’s November 2025 Connected Shoppers Report, 39% of consumers — and roughly 54% of Gen Z — said they use AI for product discovery, and 25% have already made an AI-assisted purchase. Merchants are understandably scrambling for anything that gets their catalog in front of those models, and llms.txt is one of the most talked-about — and most misunderstood — options on the table.

What is llms.txt, exactly?

llms.txt is a standard for a single Markdown file — served at your site’s root path, /llms.txt — that lists your most important pages in a clean, machine-readable format so large language models can find and summarize them quickly. The official specification describes it as a proposal “to provide information to help LLMs use a website at inference time” — in other words, at the moment an AI is answering a shopper’s question.

The idea came from Jeremy Howard, co-founder of Answer.AI and fast.ai, who published the proposal on September 3, 2024. The reasoning is practical: as the spec puts it, “large language models increasingly rely on website information, but face a critical limitation: context windows are too small to handle most websites in their entirety.” A full Shopify store — hundreds of products, collections, policies, and blog posts, wrapped in theme code and JavaScript — is far more than a model wants to read. llms.txt is meant to hand the model a shortlist instead.

The format is deliberately simple. A valid llms.txt contains:

An H1 with your store’s name. This is the only required part.
A blockquote summary — one short paragraph describing what your store sells and who it’s for.
Optional detail sections in plain Markdown.
H2-headed “file lists” — sections of [link name](url) entries pointing to your key pages (collections, best-sellers, shipping and returns, about).

It sits at the root of your domain by the same convention as robots.txt and sitemap.xml, so a model always knows where to look. You may also see a companion llms-full.txt, which inlines the actual text of those pages into one big file rather than just linking to them — think of llms.txt as the table of contents and llms-full.txt as the whole book. (Shopify has since added a third, /agents.md, which it now treats as the default — more on that below.)

Not sure whether AI assistants can even see your store yet? Run the free AI Visibility Checker — it scans your Shopify store in about 60 seconds and returns a 0–100 score across crawler access, structured data, and commerce readiness. No login required.

Does llms.txt actually work yet?

Here is the honest answer merchants rarely get: as of mid-2026, there is no confirmed evidence that any major AI assistant uses llms.txt to decide what to recommend. The file is a well-intentioned proposal that adoption has outrun. That doesn’t make it worthless, but it does mean you should treat it as a low-cost hedge, not a growth lever.

The most-cited skeptic is Google’s John Mueller. Asked about the file, he compared it to the discredited keywords meta tag in April 2025:

“AFAIK none of the AI services have said they’re using LLMs.TXT (and you can tell when you look at your server logs that they don’t even check for it). To me, it’s comparable to the keywords meta tag — this is what a site-owner claims their site is about.”

He hasn’t softened since. In June 2026 he called llms.txt “purely speculative for now,” noting the file has existed for years, “yet none of the AI systems use it.” Google’s Gary Illyes went further at a July 2025 Search Central event, stating Google doesn’t support llms.txt and isn’t planning to.

Server-log studies back the skepticism. When researchers actually watch who requests the file, almost no AI answer engines show up:

Study (date)	What it measured	What it found
Ahrefs — 137,210 domains (June 2026)	Whether anything fetches llms.txt	28% of domains publish one, but 97% of those files got zero requests in May 2026; of the requests that did happen, only ~19.5% came from AI tools
WISLR — 48-day server-log study (March 2026)	AI-bot requests for `/llms.txt`	Zero hits from GPTBot, ClaudeBot, PerplexityBot, or any AI crawler across 12,099 bot requests
Flavio Longato — Adobe Experience Manager logs (June 2026)	Who requests `/llms.txt`	Verifiable LLM crawlers were just 1.1% of 22,494 requests to the file

There is a nuance worth knowing. The one place llms.txt sees real traffic is agentic developer tooling, not shopping assistants. In the Ahrefs data, Anthropic’s coding agent, Claude-Code, out-fetched every AI retrieval bot, every AI assistant, and every AI training crawler. So if you build for developers, the file has an audience. For a typical Shopify store hoping to appear in ChatGPT’s shopping answers, that audience isn’t your customer.

Ahrefs summarized the state of play bluntly: “Adoption has been driven by speculation that AI platforms may start consuming the file, rather than by any confirmation that they do… If you want to show up in AI search, there are more reliable ways to improve your visibility than this file.” Those more reliable signals — crawler access, server-rendered content, and structured data — are the ones we cover in our guide to why most Shopify stores are invisible to AI shoppers.

Should you add an llms.txt to your Shopify store?

For most stores, the sensible answer is yes — but only as a cheap insurance policy, and only if you keep it accurate. Publishing one costs almost nothing and positions you if answer engines adopt the standard later. Betting your AI-visibility strategy on it would be a mistake.

Adoption is climbing fast even without proof it works. Originality.ai recorded an 8.8x jump — from 4,088 llms.txt files in June 2025 to 36,120 in May 2026 — and serious technical companies now publish one, including Anthropic, Stripe, and Cloudflare. That momentum is exactly why the file feels mandatory. It isn’t. Here’s the realistic split:

What an llms.txt can do for a Shopify store today:

Give agentic tools and a handful of crawlers a clean, curated map of your key pages.
Cost you essentially nothing to publish.
Serve as a low-risk hedge if ChatGPT, Perplexity, or Gemini start honoring it.

What it does not do today:

It does not guarantee — or even reliably influence — whether AI assistants recommend your store. None have confirmed they read it.
It does not replace the signals that demonstrably drive AI visibility: open crawler access, server-rendered product facts, and complete structured data.
It does not maintain itself. A stale file that lists discontinued products or old prices actively misleads any model that does read it.

As one widely shared Shopify Community guide put it, an llms.txt “does not guarantee your products will be recommended. What it does is remove friction,” and “a stale file is the real failure mode.” Or, in the words of Flavio Longato: “Create llms.txt because it is cheap and Google is now looking at it, not because it will get you cited in ChatGPT today.” Add it — then spend the bulk of your effort on the fundamentals that AI engines actually weigh, which we break down in From SEO to GEO.

The short version of adding one (and why it goes stale)

Getting an llms.txt live on Shopify used to be fiddly, because Shopify historically didn’t let merchants upload files to the domain root. The classic workaround was to upload the file under Content → Files and create a URL Redirect (Online Store → Navigation → URL Redirects) from /llms.txt to the file’s CDN URL.

That changed in mid-2026. Shopify began serving these files natively — breaking those custom redirects — and on May 28, 2026 shipped customizable Liquid templates for /agents.md, /llms.txt, and /llms-full.txt, with /agents.md as the canonical default. You can now control the output through theme templates (templates/llms.txt.liquid) under Online Store → Themes → Edit code.

The catch is the same one that sinks most llms.txt efforts: maintenance. The moment you launch a collection, retire a product, or change prices, a hand-built file drifts out of date — and a wrong map is worse than no map. Doing this by hand, and keeping it correct, is a recurring chore almost no merchant sustains. (A detailed, click-by-click setup walkthrough is a topic of its own; this guide is about understanding the file, not the full tutorial.) That maintenance problem is precisely where automation earns its place.

How Kedra AI Index handles llms.txt

Kedra AI Index is a free Shopify app built to handle the whole AI-visibility layer — including llms.txt — so you get the benefit without the upkeep. Instead of hand-editing a Markdown file and praying it stays current, the app treats your llms.txt as a living output of your real catalog.

Here’s what that looks like in practice:

Auto-generated, auto-refreshed llms.txt. Kedra AI Index publishes a live /llms.txt on your domain and rewrites it automatically as products, collections, and prices change — so the file never goes stale, which is the single most common failure mode.
Per-bot control. Choose exactly which crawlers can read your store — ClaudeBot, GPTBot, PerplexityBot, GeminiBot, and more — with a single toggle each, instead of hand-editing robots.txt.
An AI Readiness Score with clear fixes. You get a plain checklist of what’s helping and what’s hurting your discoverability — no guesswork about whether your setup is actually working.
IndexNow auto-submission. When your catalog changes, the app pings supporting engines so updates are noticed faster.
Revenue attribution by AI source. See which assistant — ChatGPT, Perplexity, Claude, or Gemini — referred each session and order, so you can measure AI as a real channel, not a mystery.

Crucially, the app doesn’t stop at llms.txt — it reinforces the signals that the research above shows actually move the needle: crawler access, structured data, and clean, crawlable content. That’s the honest way to use llms.txt: publish it cheaply, keep it accurate automatically, and invest the real effort where AI engines are proven to look.

Install Kedra AI Index free and let it generate and maintain your llms.txt while it works on the visibility signals that matter more.

Frequently asked questions

Is llms.txt the same as robots.txt?

No. robots.txt tells crawlers which pages they’re allowed to access, and it’s widely honored. llms.txt tries to tell AI models which pages matter most and summarize them — but it isn’t yet honored by major AI assistants. They sit at the same root location and follow a similar convention, yet do different jobs, and only one currently has confirmed real-world effect.

Does ChatGPT or Perplexity read my llms.txt?

There’s no confirmed evidence that they do. As of mid-2026, no major AI company has stated it uses llms.txt to generate answers, and server-log studies from Ahrefs, WISLR, and others show almost no AI answer-engine crawlers requesting the file. Anthropic’s coding agent is the notable consumer — useful for developer sites, not for typical retail discovery.

Will adding llms.txt hurt my store?

No — publishing a file has no downside for your Shopify SEO or performance, and it’s a reasonable hedge. The only real risk is a stale file: if it lists deleted products or outdated prices, any model that does read it gets wrong information. That’s why keeping it automatically in sync with your catalog matters more than simply having one.

Does Shopify support llms.txt natively?

Yes, since mid-2026. Shopify now serves /agents.md, /llms.txt, and /llms-full.txt and lets you customize them with Liquid theme templates, with /agents.md as the default. Native support handles hosting, but it doesn’t solve maintenance — keeping the content accurate as your catalog changes is still on you unless you automate it.

What should I do instead of relying on llms.txt?

Focus on the signals AI engines demonstrably use: make sure crawlers like GPTBot and ClaudeBot can reach your store, serve product facts in server-rendered HTML, and complete your Product and Offer structured data. Start by measuring where you stand with the free AI Visibility Checker, then fix the highest-impact gaps first.

The bottom line

llms.txt is a genuinely good idea that the AI ecosystem hasn’t caught up to yet. It’s cheap to publish, backed by a clean spec, and adopted by serious companies — but no major AI assistant has confirmed it reads the file, and the server logs agree. For a Shopify merchant, that makes it a sensible insurance policy, not a strategy: add one, keep it accurate, and put your real energy into the crawler access, structured data, and crawlable content that AI engines actually reward.

The smartest first move is to measure, not guess. Run the free AI Visibility Checker to see your store’s 0–100 AI-readiness score, then install Kedra AI Index to generate and auto-maintain your llms.txt and keep every signal that matters green — so when AI assistants recommend a store in your category, it’s yours.

What Is llms.txt? A Shopify Merchant's Guide

On this page