Parcel Perform is now Perform.AI. Read More, Parcel Perform is now Perform.AI.
perform.ai logo
UCP Readiness Checker

Is your store visible to ChatGPT, Perplexity and Gemini?

Shoppers increasingly start in an AI assistant rather than a search box. The stores an agent can read, price and buy from get recommended — the rest are passed over silently.

Free · Results in 60 seconds or less

[UCP checker] hero

What Every Scan Covers

147

checks run against the published UCP and Agentic Commerce specs

10

dimensions, grouped into four buyer-facing questions

≤60s

from entering a domain to a finished report

How It Works

Get your score in three steps

The whole thing runs from a domain name, in 60 seconds or less.

Enter your store URL

No integration, no setup, nothing to install. Just your domain.

We read your store the way an agent would

We fetch what an AI shopping agent fetches — your manifest, robots.txt, product pages and any service endpoints you declare — and run 147 checks against the published UCP and Agentic Commerce specifications.

Get your free readiness report

A score out of 100, the count of issues blocking a sale outright, and a plain-language breakdown of what to fix first — no technical knowledge needed.

Scored against the assistants your customers already use

ChatGPT

Perplexity

Google Gemini

Microsoft Copilot

Claude

Sample Report

See how AI engines score your store

Your report shows a readiness score out of 100, a score for each of the four categories, and every finding in plain business language — no technical knowledge needed.

A score you can act on

One readiness score out of 100, split across the four questions a shopper cares about, with the issues blocking a sale counted separately so progress is visible even when the number holds.

  • Score out of 100, with each category scored against its own weight
  • Blocking issues counted and named, not buried
  • Every finding written in business language
The report summary: four category cards scoring 87, 76, 58 and 72 per cent, each with a plain-language read and a pass or fail chip per dimension.

Findings your engineers can use

Each finding names the exact parameter it read — the field, header or endpoint — so the person fixing it knows where to look without translating from marketing copy.

  • The exact field or endpoint behind every finding
  • Prerequisites named in plain English, never as test codes
  • Untested items marked as untested, not counted against you
The Store Security dimension expanded, each finding written in plain language with the exact parameter it read beside it, such as Strict-Transport-Security.

AI search is the new front page for e‑commerce

AI commerce readiness is how well your store's signals — delivery promises, product data, trust indicators — equip AI assistants to recommend you. Unlike SEO, it's not about keywords. It's whether ChatGPT, Perplexity and Gemini have enough confidence in your store to send a shopper your way.

78%

of AI shopping traffic flows through ChatGPT alone

3×

growth in AI-driven shopping queries in the past 12 months

Weeks

not months — how fast AI recommendation rankings solidify

What Gets Scored

10 dimensions. One readiness score.

Most readiness tools stop at checkout. We score the whole journey — including what happens after the sale, where delivery promises and order tracking decide whether an agent recommends you again.

Store Identity

Does your store publish a valid UCP manifest at /.well-known/ucp, telling AI agents who you are and what you support?

AI Crawlability

Can AI shopping agents like GPTBot, ClaudeBot, Gemini and PerplexityBot reach and browse your store?

AI-Readable Product Catalog

Can an agent read your products in machine-readable form — titles, prices, availability — rather than guessing from the page?

AI Checkout

Can an agent open a checkout session, update it, and complete the purchase programmatically?

AI Payments

Do you declare which payment methods you accept, and how an agent should structure a payment?

Feature Coverage

Are checkout, fulfillment, discount and order capabilities declared with valid specification and schema URLs?

Order Lifecycle

Can an agent look up an order, read its event history, and see refunds, returns and cancellations?

Fulfillment & Delivery Data

Can an agent tell a shopper when their order will arrive, and offer more than one delivery method?

AI Connection Quality

Do the service endpoints you advertise — REST, MCP, A2A, embedded — actually respond when an agent calls them?

Store Security

Is everything served over HTTPS with the redirects and security headers the protocols require?

Frequently asked questions

Everything you need to know about getting your store ready for AI agents.

Find out where you stand

One domain, 60 seconds or less, no setup. You'll know what an AI agent can and can't do with your store.

Layered glass panels labelled UCP and ACP, the two protocols this checker scores a store against