# llms.txt: Shopify AI Discovery File 2026
> What is llms.txt, why it can help AI discovery, and why Shopify merchants should prioritize HTML, schema.org, sitemap, internal links and proof first.
- Canonical HTML: https://verityscore.io/en/kb/llms-txt/
- Markdown alternate: https://verityscore.io/en/kb/llms-txt.md
- Language: en
- Content type: kb
- Published: 2026-02-26
- Updated: 2026-04-28
- Tags: llms-txt, ai, crawlers, guide
- Audit zone: llms.txt
## What is llms.txt?
Figure 1 - robots.txt vs llms.txt: two complementary files for AI
The `llms.txt` file is an emerging convention that lets website owners publish a compact **AI discovery map** for their most important content. Placed at the site root (`https://your-store.com/llms.txt`), it provides a structured summary of your business and a list of priority links.
It complements robots.txt and sitemap.xml. In practice, it should point AI systems toward strong canonical pages; it should not carry the whole visibility strategy by itself.
Important distinction: `agent-card.json` is a capability discovery file, while `llms.txt` is a content orientation file. In Verity Score's observed logs, `/.well-known/agent-card.json` is crawled more consistently by major AI operators than `/llms.txt`. Shopify merchants should therefore treat agent-card as a serious agentic discovery surface, and llms.txt as a useful supporting index.
### Who's adopting llms.txt?
Industry adoption accelerated significantly in 2025-2026. A survey by SE Ranking published in November 2025 analyzing nearly 300,000 domains found an adoption rate of **10.13%** - meaning fewer than 1 in 10 websites currently publish an llms.txt. Important: the same study measured no statistically significant effect of llms.txt on AI citations (methodology: Spearman + XGBoost + SHAP). The low adoption still represents a competitive opportunity for early movers on structured discovery. Major organizations now publishing llms.txt include **Anthropic (Claude documentation), Cloudflare, and Stripe** (source: llmstxt.org). Mintlify rolled out llms.txt across all docs sites it hosts in late 2025, bringing thousands of documentation sites - including Cursor - into compliance overnight.
Note: John Mueller (Google) publicly confirmed (June 2025, restated in 2026): *"no AI system currently uses llms.txt"*. The file remains useful as a structured discovery index, but it's not a measurable AI visibility lever in Q1 2026. The real levers remain catalog quality, schema.org structured data, and product feed completeness.
### Google's AI Optimization Guide (May 16, 2026) confirms the position
Google has now officially published its [AI Optimization Guide](https://developers.google.com/search/docs/fundamentals/ai-optimization-guide), which explicitly rejects several "GEO hacks" promoted as universal by some vendors:
| What Google says it does NOT consume | Guide position |
|--------------------------------------|----------------|
| `llms.txt` | "AI text files" are not used by Google's AI systems |
| Markdown alternate (rel="alternate" type="text/markdown") | Same, listed as special markup with no effect |
| Custom "AI-flavored" schema.org markup | Unnecessary, standard structured data suffices for rich results |
| Artificial content chunking | Ineffective, Google understands long multi-topic pages |
| "AI-rewriting" with long-tail keyword variations | Ineffective, AI systems understand synonyms |
| Artificial brand mentions across the web | Blocked by anti-spam systems |
And what Google does confirm matters:
- Technical indexability (crawl budget, properly rendered JS)
- Semantic HTML without perfectionism
- Standard schema.org for rich results (which feed AI Overviews)
- Merchant Center + Google Business Profile
- Universal Commerce Protocol (UCP) **explicitly cited** as emerging agent surface
- 3 surfaces browser agents inspect: **screenshot (visual rendering), DOM structure, accessibility tree**
### What this means for your llms.txt strategy
llms.txt remains a valid lever, but **for Perplexity, Claude, and agent-readiness tooling (Cloudflare's isitagentready.com)** : not for Google AI Overviews. Verity Score now tags every audit signal by target engine: you see immediately whether a recommendation applies for Google, ChatGPT, Perplexity, Claude, or is universal. The "Which engine cares about each signal" panel in your audit displays this doctrine with the official citations on hover.
**Practical takeaway**: publishing llms.txt won't help your Google AI Overviews, but can improve your visibility on Perplexity and Claude, and your score on agent-readiness tools. That's not wasted effort : it's effort to scope.
The specification now defines three variants merchants can use:
- **`/llms.txt`** - links and descriptions of key pages (the standard)
- **`/llms-medium.txt`** - a single markdown page with core content
- **`/llms-full.txt`** - complete content dump with all guides, policies, and product info
For Shopify merchants, the standard `/llms.txt` with links is usually sufficient. The more important work is making the linked pages themselves crawlable, structured, sourced and consistent.
### Why freshness still matters
Research shows **50% of content cited in AI search responses is less than 13 weeks old** ([GEO 2026 State Report, 2026](https://devtripathi.in/blogs/geo-2026-ai-search-state-report/)). Freshness matters first on the canonical pages: product pages, collections, policies, FAQ, proof pages and blog guides. The llms.txt file should reflect those updates, but it is not the source of truth.
## Why it matters for AI
When an LLM visits your store to answer a query ("what's the best anti-aging serum?"), it needs to quickly understand:
- Who you are
- What you sell
- Why your products are reliable
Without llms.txt, the LLM can still discover your content through HTML links, sitemap.xml, schema.org, search indexes and external citations. That is why those surfaces must be clean first.
With llms.txt, you add a curated shortcut: a compact list of the pages you most want agents to inspect.
## Key questions to ask yourself
To know if your llms.txt is effective, check:
- **Does it exist?** An `llms.txt` file at the root of your site
- **Is it well-structured?** Title, description, links organized by section
- **Do the links work?** Each URL points to an existing page
- **Does it cover the essentials?** Your collections, policies, and key pages are represented
Verity Score automatically checks the presence and quality of your llms.txt.
## Recommended Structure
```markdown
# Your Store Name
> Short description (1-2 sentences) of what you do.
## Products
- [Category 1](https://your-store.com/collections/category-1)
- [Best-sellers](https://your-store.com/collections/best-sellers)
## About
- [Our story](https://your-store.com/pages/about)
## Policies
- [Shipping](https://your-store.com/policies/shipping-policy)
- [Returns](https://your-store.com/policies/refund-policy)
```
### Best Practices
1. **Be concise** : The LLM doesn't need your entire catalog. Guide it to key pages.
2. **Prioritize** : Put the most important collections first.
3. **Include policies** : AI shopping agents systematically check shipping and returns.
4. **Keep updated** : If you add a new collection or important content, update llms.txt.
## How to Add on Shopify
### Option 1: Via a Shopify page (recommended)
1. Create a "llms" page in Admin > Pages
2. The URL will be `https://your-store.com/pages/llms`
3. Paste your llms.txt content in plain text format in the editor
4. Add a link in the `` of your theme (file `theme.liquid`):
```html
```
Note: Shopify doesn't natively allow serving an arbitrary `.txt` file at the domain root. The `/pages/llms` URL is the simplest method and requires no technical configuration.
### Option 2: Via a CDN or proxy (exact `/llms.txt` URL)
If you use a CDN or proxy in front of Shopify (Cloudflare, Vercel, Netlify), you can configure a rewrite rule to serve an `llms.txt` file directly at the domain root. This is the ideal method if you want the exact `/llms.txt` URL conforming to the specification.
## Examples for Different Sectors
### Cosmetics / Beauty
```markdown
# Belle Peau | Organic Skincare
> Organic face care, made in France. Cosmos Organic certified formulas.
## Featured Products
- [Anti-aging serum](https://bellepeau.com/products/anti-aging-serum)
- [Moisturizer](https://bellepeau.com/products/moisturizer)
- [All face care](https://bellepeau.com/collections/face-care)
## Certifications
- [Our labels](https://bellepeau.com/pages/certifications)
```
### Fashion
```markdown
# Urban Style | Sustainable Streetwear
> Eco-friendly streetwear, manufactured in Europe.
## Collections
- [New arrivals](https://urbanstyle.com/collections/new)
- [Best-sellers](https://urbanstyle.com/collections/best-sellers)
- [Collaborations](https://urbanstyle.com/collections/collabs)
## Sustainability
- [Our approach](https://urbanstyle.com/pages/sustainability)
```
## Resources
- [llms.txt specification](https://llmstxt.org/)
- [Verity Score Guide : Agentic Readiness](/en/kb/ai-buyer-score)
- [Agent Readiness Score: the 11 /.well-known/ files to publish in 2026](/en/blog/well-known-agent-ready) : full guide with exclusive Verity data on which AI bots actually crawl your agent discovery endpoints
---
*Check if your store has an llms.txt: [free audit](/en#audit).*
## FAQ
### What is an llms.txt file and what does it do?
llms.txt is an emerging discovery file that gives AI systems a structured summary of your business and links to your most important pages. It is useful as a reading map, but it does not replace canonical HTML, schema.org, sitemap coverage, internal links or product proof.
### How do I add llms.txt to my Shopify store?
Shopify doesn't support serving .txt files at the domain root natively. The simplest method is creating a 'llms' page in Admin → Pages and linking it in your theme's head tag. If you use Cloudflare or a CDN, you can set up a rewrite rule for the exact /llms.txt URL.
### What should I include in my llms.txt file?
Include your store name, a short description, links to your main collections, best-sellers, policy pages (shipping, returns), and any key content pages (about, certifications, FAQ). Keep it concise. Guide the AI to your top 10-20 most important pages.
### Is llms.txt required for AI visibility?
No. llms.txt is optional and should be treated as a secondary discovery layer. It can help orient agents toward important pages, but observed Verity Score logs show agent-card.json is consumed more by major AI crawlers. The primary visibility levers remain crawlable HTML, structured data, product feed quality, policies, reviews, citations and internal links.
### What is the difference between robots.txt and llms.txt?
robots.txt tells crawlers what not to access. llms.txt provides an optional reading map to important content. They are complementary, but llms.txt does not replace sitemap coverage, canonical HTML, schema.org or agent-card.json.
## Sources
- [llms.txt - Standard proposé](https://llmstxt.org) (official)
- [GEO 2026: The State of Generative Engine Optimization Right Now](https://devtripathi.in/blogs/geo-2026-ai-search-state-report/) (industry)
- [Content Freshness & AI Citations Guide (2026)](https://www.qwairy.co/blog/content-freshness-ai-citations-guide) (industry)