Free toolRuns on our server; one fetch per checkProtocol v1.0Updated

Can AI shopping answers read your products?

Enter your store's address. You get a readiness report that shows what AI crawlers can reach, what your product pages say in structured data, and the exact fix for each gap.

Last reading
None published. The reading is the one you run; results are not collected into a figure.
Protocol
Scanner 1.0 (lib/config.ts): robots.txt, one sitemap, the homepage and up to 3 product pages, fetched without JavaScript inside a 20 second budget, 3 MB per page, identifying itself as OptiVis-Labs-Shelf-Check.
Verified
84 of 84 unit tests passed at build . How this is verified.
Failure state
A scan that cannot complete shows the reason in the form and produces no score. A page cut off at the size limit is reported as truncated, never softened.
Receipt
/api/scan?url= returns the same report as JSON. The print stylesheet is the download.
Method
Every check, weight and source

The same pages a crawler reads, checked one by one

  1. 01

    Read the rules

    We read your robots.txt to see which AI search crawlers may visit, then find your sitemap and the product pages it lists.

  2. 02

    Load the pages

    We load your homepage and up to 3 product pages without running JavaScript, the way many AI crawlers do. Shopify stores are detected and handled directly.

  3. 03

    Report each finding

    Every finding shows the URL we inspected, when, what we observed, how confident we are, and the exact change to make.

Six areas, weighted by how much they matter for product answers

Each area gets its own score out of 100. The overall score is the weighted average.

AreaWhat we look atWeight
Crawler accessrobots.txt rules for OAI-SearchBot, ChatGPT-User, PerplexityBot, Claude-SearchBot, Googlebot, Bingbot and others25
Product dataProduct structured data: name, image, description, brand, price, currency, availability, GTIN or MPN; business schema on the homepage30
Server renderingWhether the product name and price are in the HTML the server sends, before any JavaScript runs15
Page signalsTitles, meta descriptions, canonical tags, OpenGraph tags, noindex and nosnippet directives, image alt text15
DiscoveryA sitemap that lists product pages; llms.txt is reported but not scored10
Response speedTime to first byte, total time and HTML size, as a rough indicator only5

Read the full methodology, including the sources each check is based on.

Readiness, not appearance

This measures readiness, not appearance. Appearance in AI answers depends on factors no outside tool can see.

A high score means the basics are in place: crawlers are allowed in, product facts are stated in a form machines read reliably, and the content does not depend on JavaScript. It does not tell you how often ChatGPT, Google or anyone else shows your store, and no outside tool can tell you that.

Light on your store, light on ours

  • At most 4 pages and 5 small files per scan, each request limited to 8 seconds and 3 MB.
  • Our scanner follows your robots.txt for its own requests and identifies itself as OptiVis-Labs-Shelf-Check.
  • Results are cached for one hour per store. Each visitor can run 5 new scans per hour.
  • No email or account is needed to see results. What we store.

Need the fix done? OptiVis Marketing does this work for stores.

What information does AI Shelf Check collect and retain?

OptiVis AI Labs offers AI Shelf Check as a free AI-shopping readiness checker. For questions about information submitted to the tool, consult its privacy notice and contact OptiVis LLC at info@optivis.dev. The notice is the reference for data collection, use, and retention when deciding what to submit.

AI Shelf Check privacy notice

Facts reviewed October 6, 2026.