Free agent readiness scanner for online stores: Could an AI shopping agent buy from your store?

Paste your store’s address. In about five minutes we walk it like an AI agent and show you where it would stop.

Free, and you need no account to scan. The full report is free too: it opens after you enter your email and the six-digit code we send to it.

GPTBotClaudeBotOAI-SearchBotGooglebotAmazonbotChatGPT-UserClaude-UserPerplexityBotGoogle-ExtendedAmzn-SearchBotmeta-externalagentApplebotchatgpt.comClaude-SearchBotPerplexity-UserGoogle-AgentAmzn-Usermeta-externalfetcherApplebot-ExtendedbingbotBytespiderCCBot

What is agent readiness?

Agent readiness is whether an AI shopping agent can find your store, compare what you sell and buy from you. Agentic commerce is AI agents doing that shopping on a person’s behalf: the agent does the browsing, and the person still pays.

Your agent readiness score is the Ready score in the report: a number out of 100, read from your own pages, with the step where each point was lost. When the evidence is too thin to support a number, the report leaves the score out and says why.

The path an agent takes

Each agent DISC-001 looks for in your robots.txt comes in through the door and goes where its intent takes it. Crawlers that train a model or index for search read the shelves. Agents browsing for a person look over the display table, and the rare one that came to buy heads for the till.

Trains a model 8
Indexes for search 7
Browses for a person 6
Tries to buy 1

Hover or tab to a figure to see who it is. The arrow keys step between them.

What comes back

01

Where an agent would stop

The first step an agent could not get past on your shop, and what the scan read there.

02

Ready score

A score out of 100 for whether an agent could find your products, read prices and stock, and check out as a guest. We read it from your own pages and show the step where each point was lost.

03

What each check read

For every check, the page or markup it read on your shop, or what it could not find. You can forward this part to the team that fixes it.

04

The fix list

Each finding says who fixes it, how much effort it takes and which check produced the evidence, so you can hand it straight to that person.

A second score, Safe, shows whether your shop can tell a real agent from a spoofed one. You need to verify your domain first, because it reads things we should only read for the owner.

Name up to two competitors when you start a scan, and the report shows their Ready score beside yours, measured by the same checks on their shop.

51 checks on a passive scan, along 4 steps

What we check

We group the checks by the step of the path they read. Each id is the one your report prints beside a finding, and each links to a page that explains the check. The Safe active probes are listed too; they run only after you verify ownership. So is MEAS-001, which needs your own traffic logs, so a passive scan does not run it.

22 named agents from 10 vendors, taken from the list DISC-001 checks your robots.txt against. Your report says whether your shop lets any of them in.

Hover over a cart to see whose it is, or tab to the belt and step along it with the arrow keys.

Find

An agent has to find your shop and your products without a person pointing the way.

Compare

It has to read price, stock, delivery and returns as text it can trust.

Buy

It has to reach the payment step as a guest, with nothing in the way.

After the order

It has to track the order, start a return and reach support.

Support contact structured

An AI shopping agent walking the aisles of a store, with its route traced in green

The limits we keep

We scan shops whose owners did not ask us to. The limits that make that fair are written down and enforced in the crawler’s code, and two lines in your robots.txt keep us out. Our reports tell you which AI crawlers your robots.txt blocks, and you should be able to block ours as easily.

1

MCP request, at most

Per scan, and only to the MCP server your UCP profile declares, wherever it is hosted: a read-only tools/list call that changes nothing. We check the specification and schema links in that profile only on ucp.dev, shopify.dev and pay.google.com.

2

requests a second

We pace requests to each host with a token bucket, so a scan of your shop is quieter than one person browsing it.

Both

count a rendered page

Each scan has a fixed request budget set in the crawler’s code, and the crawler refuses any request past it before it reaches the network. A page we open in a real browser is charged against that budget and the rate limit exactly like a plain fetch.

First

we read robots.txt

We read it at the start of every scan, before anything else. If it refuses our crawler, we fetch your home page once so the report can say “blocked” instead of “saw nothing”, and then we stop.

30

days, then deleted

Screenshots and captured page data expire on a storage lifecycle rule, so nobody has to remember to delete them.

One

user agent, never disguised

We send the same name on plain requests and on the pages we render. We never rotate it, and we never dress the crawler up as an ordinary browser to get past a rule you set for it.

Or read the whole thing on /bot

What we cannot tell you

no score

When the evidence is thin, there is no score.

The report keeps the space where the score would be and prints the reason it was withheld.

skipped

A check that could not run says so.

When bot management answers instead of the shop, or a page never loads, the check is skipped with the reason attached. We do not guess its result or average it into the score.

no logos

We have no customers to show you.

Storewalk is new. Everything on this page comes from our own test store or a real scan of a real shop, and we have no customer logos to put here yet.

How this scan differs from isitagentready and Shopify’s audit

When we last read their pages, isitagentready, which its footer credits to Cloudflare, checked whether a site publishes the files and protocols agents look for, such as AI rules in robots.txt, Markdown versions of pages and an MCP server card. None of its checks read products, the cart or checkout. Shopify’s product page audit for agentic commerce took one product page and read its structured data and whether robots.txt let the main AI crawlers in.

Storewalk walks your store along the path an agent takes, from the home page and your products to the cart and checkout pages. When bot protection answers instead of your store, or a page never loads, the check says SKIPPED with the reason, and it stays out of the score instead of being guessed.

Each of these is a fair AI agent readiness checker for the question it was built to answer. Alternatives to isitagentready sets them beside the other tools we found, ours included.

Questions about the scan

Is the agent readiness scanner free?

Yes. A scan costs nothing and needs no account. The first results show on screen, and the full report opens after you enter your email and the six-digit code we send to it. The second score, Safe, reads things only an owner should see, so it waits until you verify your domain in an account.

How long does a scan take?

About five minutes. The progress page follows the scan as it reads your store, so you can watch it work or go and make a coffee.

Which bots and standards does it check?

Your robots.txt rules for the AI crawlers and agents of OpenAI, Anthropic, Perplexity, Google, Amazon, Meta, Apple, Microsoft, ByteDance and Common Crawl. Then the files and markup an agent reads on the way to buying: llms.txt, your sitemap, Product structured data, pages served as Markdown, discovery files such as a UCP profile or an MCP server card, guest checkout, a CAPTCHA before payment, and agent payment protocols such as x402. Each check has its own page in the checklist.

Which store platforms does it work with?

Any store with a public address. The scan reads your pages the way an agent would, whatever they are built on. We also have guides with the settings for Shopify, WooCommerce, BigCommerce and Magento.

Why do some checks say SKIPPED?

A check is skipped when it could not run: bot protection answered instead of your store, a page never loaded, or the empty cart sent the scanner away from checkout. The reason is printed beside it. We do not guess the result or count it in the score, and when too much is missing the report shows no score at all and says why.

Does the scan change anything on my store?

No. It reads pages, the cart and checkout included, and adds nothing to a cart, fills in no form and creates no account. If your UCP profile declares an MCP server, it sends that server one read-only tools/list request. It reads robots.txt first, and if that refuses our crawler it opens your home page once and stops. How our crawler works has the details and the line that blocks it.

Who can assess my store for AI agent readiness?

You can, in about five minutes: run the free scan and read the report. If you would rather have a person go through the results with you and turn them into a fix plan, ask for the free consultation when the report offers one, or write to us from the contact page.

From the blog

How an AI shopping agent reads a store, the protocols it buys through, and where checkout and payment stop it.

Why AI shopping agents can't buy from most stores yet
Five technical walls stand between an agent and a completed purchase: reading the page, proving who it is, finding what the store offers, getting through checkout, and paying on someone’s behalf.
UCP explained
The Universal Commerce Protocol: the profile a store publishes at /.well-known/ucp, the capabilities it declares there, and the checkout flow from a new basket to a placed order.
How an agent pays without seeing your card
Tokenised and delegated payment: Stripe Shared Payment Tokens, ACP’s delegated payment, what Adyen’s and PayPal’s documents describe, Visa’s Trusted Agent Protocol and AP2’s mandates.
Alternatives to isitagentready: tools that test whether agents can use your store
What isitagentready checks and what it leaves out, then ten other tools and test suites, Shopify’s own audit among them, each described from its own pages, with the question each can answer. One of them, Storewalk, is ours.

Guides by platform: Shopify, WooCommerce, BigCommerce, Magento.

All articles All guides

Want help fixing what the scan finds?

Every full report comes with a free consultation: we go through what the scan found with you and tell you what to fix first. Ask for it with one click on your report, and we'll email you within one business day.

Or pick a time now

See the step where an agent would stop in your store.

All we need is your store’s address, and you do not need an account. You get a score for each step, the findings, and a list of who fixes what.

Scan a store Or read what our crawler does, and how to block it