// we don't guess — we read the same bytes the bot does
Shoppers ask ChatGPT, Perplexity, and Google AI what to buy. Those agents don't see your design — they parse your source. This tool reads the exact signals they read and tells you whether they can find, understand, and recommend your product — then what to fix. Free, no signup, results in seconds.
AI recommends the best option, not just an adequate one. Paste 1–2 rival product URLs to see how your signals stack up.
The eight things AI shopping agents read before they'll recommend a product, all checked from public pages with no login and nothing stored.
Product markup with price, currency, stock, brand, image, a GTIN or SKU and breadcrumbs, plus what Google's merchant listings and OpenAI's product feed ask for: a name under 150 characters, a real description, one Offer, standard availability values and valid JSON-LD.
Star rating markup, review count and ratings shown on the page, and a flag for rating markup the page's reviews don't back up, which Google penalizes.
Store name markup, return and shipping terms written on the product page or in the markup (footer links don't count), and the return, shipping, privacy and terms pages agents open to answer questions.
Size and dimensions, materials, care, and fit or compatibility, field by field, plus specs, an FAQ and a real description in the page HTML. Digital products skip the physical details.
The markup's price, stock and SKUs against the store's live product data on Shopify (or the price and buy button shown elsewhere), every variant's own SKU, barcode and photo, valid GTIN check digits, and clear option names.
Machine-readable price and currency, a canonical URL, Open Graph product tags and a page title of 30 to 60 characters.
robots.txt read the way the bots read it, for 12 search and AI bots including Googlebot, Bingbot, OAI-SearchBot and PerplexityBot, plus noindex tags, the canonical tag, the sitemap, and firewalls that turn AI bots away.
Whether a browsing agent can find the buy button in the page HTML, read the option labels, and add to cart without your scripts running.
It's how well a product page and its store data give AI shopping agents (ChatGPT, Google AI Mode and Gemini, Perplexity, Copilot) what they need to read, match and recommend a product: Product structured data that matches the store's price and stock, identifiers like a valid GTIN, every variant's details, reviews, return and shipping terms, answers to buyer questions, and access for search and AI crawlers. This tool checks each one from a product URL and returns a score with specific fixes.
Eight areas: product identity (Product markup, price, image, a valid GTIN, valid JSON-LD), reviews (including rating markup the page's reviews don't back up), store trust (return and shipping terms on the product page and in the policy pages), buyer questions (size, materials, care, fit, an FAQ and a real description), price, stock and variants (the markup against the store's own data, and each variant's SKU, barcode and photo), page basics (canonical tag, title, Open Graph), crawler access and indexing (robots.txt for 12 bots, noindex, the sitemap, firewalls that turn AI bots away), and whether a browsing agent can find and use the buy button. It also maps the page to OpenAI's product feed fields.
Yes. It's free with no account and no limit. It reads the public product page you enter, plus its robots.txt, sitemap and policy pages and, on Shopify, the store's public product data. It stores nothing, and there's no AI or paid API on our side, so there's nothing to charge for.
No. The check is a fast rules-based read of the page's structured data, meta tags and text, the store's robots.txt, sitemap and policy pages, and Shopify's public product data. It doesn't query ChatGPT, Perplexity or any model, so it can't tell you the exact answer an AI would give. It tells you whether your product exposes the facts those engines rely on to recommend it.
Some stores render Product structured data in the browser with JavaScript, so it's not in the raw HTML the checker reads. If your schema loads client-side, an AI crawler that doesn't run JavaScript may miss it too. The fix is to output Product and Offer JSON-LD in the server-rendered HTML.
Yes. It works on any public product page, on Shopify, WooCommerce or a custom stack, because the core checks use schema.org markup, Open Graph tags and robots.txt. On Shopify it also reads the store's public product data, so it can check every variant against the page. The app recommendations are geared to Shopify merchants.
