Check a store
Enter a store URL. The check reads only public pages — no login, no private data. Results show what was actually found, and anything that could not be verified is labelled as unverified, never guessed.
Checking… this usually takes 10–40 seconds. The store is fetched once per check; repeat checks within 24 hours use the saved result.
How it works
- Enter the store URL. The checker fetches the store's public pages from the server.
- 5 products are sampled. On Shopify stores the sample comes from the public products feed; on other stores it comes from the sitemap. If neither is available, only the submitted page is checked and the report says so.
- You get a plain-English report. Each check shows pass, fail, warning, or unverified, with coverage percentages and the fixes that matter most.
What is checked
- Homepage reachable, title, meta description, and noindex status
- robots.txt, including the voluntary
__usage_ai-trainingsignal - llms.txt and agents.md files
- sitemap.xml
- Product structured data: price, availability, reviews, shipping and returns
- Price and stock consistency between the product page and the products feed (Shopify only)
This checks public signals only. It does not certify AI-agent readiness and does not guarantee citations, rankings, or sales.
What AI shopping agents look for
AI shopping agents read the same public pages shoppers see, but they rely on structured data, not design. If a product's price, availability, or reviews are missing from the page's structured data, an agent may skip the product, show it with incomplete information, or not recommend it at all.
Shopping agents inside ChatGPT, Google's AI Overviews and AI Mode, Microsoft Copilot, and Perplexity discover products through search results, product feeds, and structured data on the page. Shopify reports that AI-driven traffic to Shopify stores grew 8 times year over year in Q1 2026, while orders from AI-powered searches rose nearly 13 times. The surfaces are still new and mostly US-first, but the direction is clear: product data that agents can read is becoming part of being found.
Five things agents need from every product page:
- Product structured data — name, price, and availability in a format machines can parse.
- Fresh price and stock — the same values on the page, in the feed, and at checkout.
- Review data — ratings and review counts that agents can quote with confidence.
- Clear shipping and return policies — so the agent can answer cost and return questions.
- Machine-readable site files — robots.txt, a sitemap, and llms.txt or agents.md so agents can find and understand the store.
What this checker examines
Every check below is performed on the store's live public pages. Anything that cannot be fetched is labelled "could not verify", never guessed and never counted as a failure.
Product schema coverage
The share of sampled products with valid Product structured data (JSON-LD). A product without it is harder for agents to understand, whatever the page looks like to a human visitor.
Price and availability
Whether each sampled product exposes a price and an availability status in its structured data, and — on Shopify stores — whether those values match the public products feed. A mismatch means agents can show the wrong price or an out-of-stock item as available.
Reviews
Whether ratings and review counts appear in structured data. A store can collect reviews and still show none to agents: review apps do not always expose ratings in a machine-readable form, so this check reads what is actually on the page.
Shipping and returns
Whether shipping and return information is present in structured data, so agents can answer delivery-cost and return questions without guessing.
robots.txt and the AI-training signal
Whether robots.txt exists, and whether it carries the voluntary __usage_ai-training signal. That signal asks AI companies not to use the site's content for training. It is training-only: it does not block crawling, search indexing, or AI answers.
llms.txt and agents.md
Whether the store publishes a machine-readable file describing the site for AI systems. These files are a proposal, not a web standard, and Google Search ignores llms.txt — but other AI systems read them, so their presence is reported as a signal, not a requirement.
Sitemap, meta data, and indexability
Whether sitemap.xml exists so products can be discovered, whether the homepage has a meta title and description, and whether the homepage is blocked from indexing. A store agents cannot discover cannot be recommended.
Free check vs full audit
The free check is a sample. It shows the pattern of problems across 5 products. The full audit covers the whole catalog and ends with fixes, not just findings.
| Free check | Full audit | |
|---|---|---|
| Products checked | 5 sampled products | Every product, up to 250 in one scan |
| Price and stock consistency | Sampled products only | Whole catalog, page against feed |
| Schema, review, and shipping coverage | Percentages across the sample | Percentages across the whole catalog |
| Report | Instant, on this page | Full report, fixes applied, then re-checked |
| Price | Free, no signup | Manual quote before work starts |
To request a full audit, run the free check first, then use the fix-request form or the WhatsApp button in the report. No payment is taken on this page.
Who this checker is for
- Shopify store owners — especially stores selling to buyers in the US, where AI shopping and checkout surfaces are rolling out first.
- WooCommerce and WordPress stores — product pages are sampled from the sitemap when no products feed exists.
- Agencies and freelancers — run the free check on a client store before quoting catalog cleanup work.
If you sell through another platform, the checker still examines the homepage, robots.txt, sitemap, and any product pages it can find from the sitemap, and says plainly when only a single page could be checked.
Common issues this checker finds
- Reviews collected but not exposed. The store has a review app, yet no ratings appear in structured data. The fix is usually in the review app's settings or theme code, so ratings become readable.
- Price or stock disagrees between page and feed. The product page and the products feed show different values. The fix is to find which source is stale and sync them.
- Missing machine-readable files. No llms.txt, agents.md, or AI-training signal in robots.txt. The fix is to add the files with accurate store information.
- Thin homepage signals. Missing meta description or a sitemap that omits products. The fix is basic on-page work that also helps ordinary search.
The report names the exact products where each issue was found, so fixes start from evidence, not guesses.
Frequently asked questions
What does this checker look at?
Only public signals: whether the homepage is reachable and indexed, robots.txt and llms.txt files, the sitemap, and Product structured data (price, availability, reviews, shipping and returns) on a sample of product pages. It does not log in to the store or read private data.
Why does the free check only cover 5 products?
The free check is a teaser sample. It checks 5 products so a store owner can see the pattern of problems quickly. A full-catalog audit checks every product and is available on request.
Does a good score guarantee AI citations or sales?
No. This tool checks public signals only. It does not certify AI-agent readiness and does not guarantee citations, rankings, or sales.
How often can I check the same store?
The same store can be checked once every 24 hours from the same connection. After that the free check unlocks again, or a full audit can be requested.
Which platforms does it support?
Shopify and WooCommerce (WordPress) stores, and any online store with a sitemap. If no product feed or sitemap is available, the checker examines the single submitted page and says so in the report.
What is the __usage_ai-training signal?
It is a voluntary robots.txt signal that asks AI companies not to use a site's content for training. It covers training only: it does not block crawling, search indexing, or AI answers, and it does not undo past training.
Is my store's data stored?
Check results are cached for 24 hours so repeat checks do not re-fetch the store. The checker reads only public pages and stores no login or private data. Your contact details are saved only if you submit the fix-request form.
What happens after I request a fix quote?
Your report and details are sent for a manual review, and you receive a quote for the fixes. No payment is taken on this page. For a faster reply, use the WhatsApp button.
Can the checker read a store that needs a login?
No. The checker reads only publicly reachable pages. Pages behind a login, paywall, or block cannot be verified and are labelled as unverified, never guessed.