Check the product data AI agents read
A product data checker reads the schema.org Product markup on one product page and reports each field an AI agent reads: name, image, brand, price, currency, availability, GTIN and MPN, whether each is present and in a usable form, and why an agent needs it. Paste a product URL to see yours.
How to read the result
MISSING in red is a field whose absence a scan reports as a finding: no Product markup, no price, no stock status, or no identifier. MISSING in amber is a field an agent uses that the scan does not grade. CHECK means the field is there in a form an agent may not read, such as a price written with a currency symbol or a barcode with the wrong check digit. #
Most fixes are a setting in your platform or theme, not code: a barcode field left empty on the product, or a theme that prints the price without its currency. Forward the table to your developer if it is not.
Related checks
Agent view shows the whole page as an AI agent receives it, including the text around the markup. The missing GTIN stall explains what an identifier costs a store that does not publish one.
Frequently asked questions
What is product structured data?
A block of schema.org markup in a product page, usually JSON-LD in a script tag, that states the facts of the product in a form software reads without guessing: its name, price, currency, whether it is in stock, its barcode and its brand. Shopify, WooCommerce and most themes print one, and many leave fields empty.
Which fields does an AI shopping agent need?
A price with its currency, and whether the item is in stock, before anything else: those decide whether an agent can offer the product at all. Then an identifier, a GTIN or an MPN, so the agent can match your product with the one a shopper named. Name, brand, image and description shape how it presents the product.
Why does the GTIN matter so much?
It is the one field that says this product is the same item as the one a shopper asked for by name elsewhere. Without it, an agent matching products across stores has only the name to go on, and names differ from store to store. The check digit is validated here, because a mistyped barcode matches nothing.
My page uses microdata, not JSON-LD. Is that wrong?
No. Microdata and RDFa are valid schema.org syntax, and a scan raises no finding for them. This tool reads fields from JSON-LD only, so it says which syntax you use and does not report the fields one by one.
Is this the same as the scan?
It reads one page with the same parser and the same rules the scan uses, so the four findings it can report are the ones a scan would report for that page. A scan reads several products, your catalogue and your policies, and then walks a real purchase as far as the payment step.
What do you send to my store?
Two requests: robots.txt, then the page you entered, from our published user agent and a fixed set of static IP addresses listed on our agent page. If robots.txt asks our agent not to read the page, we do not. A result is kept for five minutes, and we cap how often any one store is checked across all our free tools, however many people ask.
See how our own agent identifies itself, and how to opt out, on our agent page.
Readable is not the same as buyable
A scan walks your store the way an AI shopper would and tells you where it stalls on the way to checkout. One free scan per store, per email address, every 30 days.