What Does Your Shopify Store Need to Be Ready for AI Agents?
Agentic commerce is here. AI agents are already browsing product catalogs, comparing prices, and making purchase recommendations for real customers. Most Shopify stores are invisible to them — not because they're bad stores, but because they're missing 3-4 specific technical requirements.
By Steve Merrill | June 29, 2026
Here's the checklist. Work through it once and you'll be ahead of 90% of merchants on Shopify.
The 12-Point Agentic Commerce Readiness Checklist
1. Shopify Catalog API Is Active and Complete
This is the main channel AI shopping agents use to access your product data. If it's returning incomplete product information — missing variants, stale prices, or broken availability signals — agents skip you. Go to Shopify Admin → Apps → Sales Channels and verify your Catalog API is enabled. Check for errors in the feed.
2. Product Schema Markup on Every Page
Schema.org Product markup tells AI what your products are. Required fields: name, description, sku, brand, offers (with price, priceCurrency, availability). Missing even one of these and AI agents may skip your page entirely. Use Google's Rich Results Test to audit your current schema.
3. Offer Schema With Real-Time Pricing
Separate from Product schema. Offer schema must include current price, currency, availability (InStock/OutOfStock/LimitedAvailability), and seller. Agents that recommend products to buyers need this to give accurate purchase information. Outdated pricing in your schema = your products getting flagged as unreliable data.
4. llms.txt File at Your Domain Root
This is the newest requirement and most stores are missing it. Create a file at yourdomain.com/llms.txt that tells AI what your store sells, your key product categories, and how you want your brand represented. Think of it as your store's resume for AI. LLMs.txt specification takes 15 minutes to implement.
5. AI-Ready Product Descriptions
Vague descriptions like "great quality for everyday use" are invisible to AI. Agents parse structured attributes: materials, dimensions, weight, compatibility, use cases. Audit your top 20 products. Rewrite descriptions to answer: what is it, what's it made of, who is it for, what problem does it solve, what size/weight/dimensions does it come in.
6. Google Merchant Center Feed Active With Zero Errors
Multiple AI shopping surfaces — including Google AI Mode — pull product data directly from Merchant Center. Log in and check for feed errors. Common blockers: missing GTINs, incomplete shipping data, price mismatches between feed and website.
7. Merchant Return Policy in Structured Format
When an AI agent is helping someone decide whether to buy, return policy is one of the first things it checks. Add MerchantReturnPolicy schema to your store. Include return window, refund type, and process. This isn't optional — it's how agents communicate trust signals.
8. Page Speed Under 2.5s LCP
AI crawlers timeout on slow stores. Shopify stores with LCP over 4 seconds are getting lower crawl priority. Run your homepage and top product pages through PageSpeed Insights. Address anything that drops you below a 70 performance score.
9. BreadcrumbList Schema on Category and Product Pages
This tells AI agents how your products are organized. An agent recommending running shoes needs to understand you sell athletic footwear, not just the single product page it landed on. BreadcrumbList schema creates that navigational context.
10. robots.txt Allows Major AI Crawlers
Check your robots.txt at yourdomain.com/robots.txt. AI shopping agents use specific user-agent strings: PerplexityBot, GPTBot, GoogleOther, Googlebot-Shopping. If your Disallow rules are blocking these, you're blocking AI discovery at the front door.
11. AggregateRating Schema on Product Pages
AI agents prioritize products with verifiable social proof. If you have reviews, surface them in structured data. AggregateRating schema needs: ratingValue, reviewCount, and bestRating. Without this, your 4.8-star product looks identical to one with no reviews when viewed by an AI.
12. Manual AI Shopping Test (The Sanity Check)
After all of the above: go to ChatGPT Shopping, Perplexity, and Google AI Mode. Search for your actual products. Document every gap between what you find and what you sell. This is your live audit. The gaps are your remaining action items.
Which Items Matter Most?
If you can only do three things this week: fix item 1 (Catalog API), add item 2 (Product schema), and write item 4 (llms.txt). These three unlock the most AI surfaces for most Shopify merchants. Everything else builds on them.
Frequently Asked Questions
How long does it take to complete this checklist?
For a Shopify store with an active developer or someone comfortable in the admin: 4-6 hours for items 1-11. The manual testing in item 12 is ongoing — plan to do it monthly.
Does Shopify handle any of this automatically?
Basic Product schema is generated automatically by most Shopify themes. But it's rarely complete. Offer schema, MerchantReturnPolicy, BreadcrumbList, AggregateRating, and llms.txt all require manual setup or an app.
What's the difference between agentic commerce readiness and regular SEO?
SEO optimizes for search engine rankings. Agentic commerce readiness optimizes for AI agent decision-making. Different signals, different data requirements. You need both, but the checklist above is specifically for AI agents that browse and recommend products directly.
Which AI shopping agents should I test with?
Start with ChatGPT Shopping (available in ChatGPT with browsing), Perplexity AI (shopping mode), and Google AI Mode. These three cover the majority of current AI-driven product discovery. Test on desktop and mobile — results vary by platform.
How often should I run this checklist?
Full checklist: quarterly. The manual AI test in item 12: monthly. Schema validation: every time you make major site changes.
Get Your Store Audit-Ready
Working through this checklist takes a few hours, but the alternative is having your products invisible while AI agents recommend your competitors. Start with the Catalog API and Product schema — those two changes alone have moved stores from zero AI citations to consistent recommendations within 3-4 weeks.
Want a done-for-you audit of all 12 items against your specific store? See how WRKNG Digital's AI commerce audit works →
