How to check if your store is visible to AI shopping agents
Three ways to find out what an AI shopping agent actually sees on your store, from a free automated grade to manual source-code checks.
AI shopping agents are starting to send buyers to stores. Before you worry about ranking in their answers, settle a more basic question: can they read your store at all? Here are three ways to find out what an agent gets when it fetches your page, fastest first.
1. Run a free automated grade
The quickest answer is a tool that fetches your live page and reports the signals an agent reads. The free ShelfGrader scan does this in about a minute: paste your store URL and you get a grade out of 100, a ranked list of what is missing, and the competitor an agent would pick instead. It checks the same things an agent checks, so you do not have to know what to look for yourself.
2. Read your own page source
You can do a rough version by hand. Open a product page, view the page source (right click, then View Page Source), and search the raw HTML for a few things:
- Your price. Search for the number. If it is not in the source, it is probably locked in an image or rendered late by a script, and an agent may not see it.
- Product schema. Search for
application/ld+jsonand"@type": "Product". If neither appears, you have no Product schema, which is the most common reason stores get skipped. - Availability and policies. Look for in-stock wording, shipping and returns in readable text, not just icons.
This tells you whether the basics are present, though it will not rank the fixes or show you how you compare to a competitor.
3. Validate your structured data
If you do have schema, confirm it is valid. Google's Rich Results Test and the Schema Markup Validator both read a URL and flag missing or malformed fields. Valid schema is what an agent trusts; broken schema can be worse than none, because it can feed an agent wrong data and get you filtered out anyway.
What to do with the result
All three methods ask the same question: are your price, availability, product details and ratings present in a form a machine can read? If not, add Product schema, move the price into the HTML, and write your policies as text. For what to do once an agent can read you, see how to get your store recommended by AI shopping agents.
If you only do one thing, run the free grade and start with whatever it flags first.
Frequently asked questions
How do I know if AI agents can read my store?
Check whether the facts an agent needs are present in your page's HTML: a machine-readable price, Product schema (JSON-LD), availability, and ratings. You can do this manually by viewing the page source, or instantly with a tool like the free ShelfGrader scan that reads your live page the way an agent does and reports what is missing.
Can I just ask ChatGPT if it can see my store?
It is a useful spot check but not reliable. Asking an assistant about your store can pull from training data, search results or a live fetch, so a confident answer does not prove your current page is readable. A direct check of your page's HTML, or a tool that fetches your live page, tells you what is actually there now.
What is a good AI-visibility score?
On ShelfGrader's 100-point grade, most stores we scan land around 57. Anything in the 80s and up means the core signals an agent needs are present. The number matters less than which specific signals are missing, since one disqualifying gap, like no readable price, can get you skipped regardless of the rest.