1. Home
  2. Tools
  3. AI agent readiness

Can AI agents read and shop your store?

Paste a store or site address. The checker fetches it the way an AI crawler would, then scores what it finds: crawler rules, llms.txt, Markdown, structured data, a real product page, Shopify's agentic checkout profile and security headers. Every gap comes with a fix you can copy.

Some links may earn us a commission at no cost to you. Drafted with AI, checked by a human. How we earn.

  • Checks the live site
  • A fix to copy for every gap
  • No sign-up
Illustration of a shop front with a checklist and a search box

Check a store or site

Live scan, usually under 15 seconds

Try one:

Only public pages are read: the homepage, one product page, robots.txt, the sitemap, llms.txt and a few standard files. Private and internal addresses are refused. Results are cached for 10 minutes, then expire.

What the checker looks at

How the score works

Each scored check carries a weight: high impact counts 3, medium 2, low 1. A pass earns the full weight, a partial pass half, a miss nothing. The score is the share of the available weight you earned.

Some things never count against you: blocking training crawlers such as GPTBot or ClaudeBot, optional agent files you have no reason to publish, and anything the checker could not verify (a timeout, for example). Store checks only apply when the site looks like a store, and the UCP and agents.md checks are only scored on Shopify, where they exist by default.

A score is a starting point, not a ranking factor. The checks that matter most for a store are the ones that decide whether an assistant can reach a product page and quote its price, stock and returns policy.

Fixing the gaps

Every gap has a short reason and a snippet to copy. "Copy all fixes as a prompt" puts them in one list you can hand to a developer or a coding assistant. To set crawler rules without writing them by hand, use the AI crawler robots.txt generator.

FAQ

What does an AI agent readiness check actually test?

It requests your homepage, robots.txt, sitemap, llms.txt and a handful of standard files the way an AI crawler does, then checks whether answer engines are allowed in, whether the content is in the HTML without JavaScript, whether a product page carries price and stock in structured data, and whether Shopify's agentic checkout profile (UCP) is reachable.

Will blocking GPTBot or ClaudeBot lower my score?

No. Those are training crawlers, and blocking them is a policy choice that does not remove you from AI answers. The score only drops if you block the bots that feed answers and search, such as OAI-SearchBot, Claude-SearchBot, PerplexityBot, Googlebot or Bingbot.

Do I need an MCP server, an A2A card or an OpenAPI file?

Only if you actually run one. They are shown as optional and never count against the score. A Shopify store already gets an MCP endpoint for shopping agents through its UCP profile.

My store is on Shopify. Which fixes can I make myself?

robots.txt (through the robots.txt.liquid template), product data such as barcodes, SKUs and alt text, policy pages, and anything in your theme. Response headers and Markdown responses are controlled by Shopify on hosted storefronts, and the report says so when it applies.

Is the scan stored?

The result is cached for 10 minutes so a repeat check is instant, then it expires. Only public pages are requested, and the checker refuses private, internal and reserved network addresses.

Why did my site score differently from another checker?

Checkers weigh things differently. This one leaves out anything it could not verify, does not penalise training-bot blocks or optional agent files, and weighs the checks that decide whether a shopping agent can find and quote a product most heavily.