UCP implementation

A shopper asks an AI assistant to find a waterproof jacket in their size, within their budget, delivered by Friday. The assistant compares options and completes the purchase. Whether your store is one of those options depends on whether an agent can read your catalog and transact with it.

The same applies in B2B. A purchasing team can instruct an agent to source screws that meet a given specification. The agent finds suppliers that publish a UCP business profile, uses the declared catalog and commerce capabilities to compare offers, terms, and availability, then places the order according to its approval rules. This only works if your catalog, pricing rules, and stock data are accurate and reachable.

On Magento, WooCommerce, Umbraco, and custom builds, that connection has to be built. We build it, test it against real purchase conditions, and hand over the code in your own repository.

Shopify is the exception. UCP is built into the platform, so the job there is different. What decides whether an agent picks your product is the catalog behind it: titles, attributes, variants, and delivery information clear enough for an agent to compare against a shopper's request. We do that work too.

The Universal Commerce Protocol is an open standard that gives AI agents one defined way to discover what a store supports and then search, check out, and place orders through it. Without a shared standard, every agent surface is a separate integration. With it, you build once. The standard itself is free; the cost sits in connecting it to the systems that already run your store.

UCP checkout is live in a limited set of markets but is continuously expanding. A store outside a live market can still be assessed and prepared now. Cleaning up your product information and making sure your inventory, pricing, and checkout systems match can take time. Starting early helps your store be ready as soon as UCP becomes available in your market.

For the full picture, including what breaks at runtime and how the specification has changed since launch, read our blog about how UCP is free, but implementing it isn't.

Some platforms offer ready-made plugins that let AI agents communicate with a store. But these plugins are built for a standard setup. They don't automatically understand how your business handles product data, volume pricing, stock, or a customized checkout. Connecting those parts is where most of the work lies.

We implement UCP as a layer over the systems already running your store rather than replacing them, drawing on years of experience in building and extending Magento, WooCommerce, Umbraco, Shopify, and custom ecommerce systems. That covers both the implementation work on platforms without native support, and the data and configuration work Shopify stores need instead.

UCP can connect your store to AI agents in more than one way. It can use standard web connections such as REST or MCP, which is a common way for AI systems to connect to tools and data. We do both sides of that connection. Read about our MCP server development, which makes your store available to approved agents, and our AI agent development, which builds agents that use it.

• Published UCP profile: A profile at /.well-known/ucp that declares one pinned version, with a self-contained profile for every supported version.
• Phased capability endpoints: Checkout and Catalog first, then Cart and Identity Linking. Further capabilities and extensions, including Permalink, Location, Fulfillment, Discounts, and Buyer Consent, are added when your business or platform calls for them.
• Identity linking: OAuth 2.0 with PKCE, scoped tokens, consent, revocation, and a guest checkout path that continues to work without account linking.
• Payment handler integration: Connection to your existing payment service provider through UCP's payment handler and token exchange model, including 3D Secure, declines, and retries. UCP does not replace your payment stack, and you remain the merchant of record, so the customer relationship and the transaction stay yours.
• Order webhooks: Signature verification, idempotency keys, retry handling, and monitoring for duplicate or delayed events.

If UCP is part of a wider store or back-office integration project, our ecommerce development and integration work can cover the surrounding systems as well.

Catalog endpoints make product data available to agents, but they do not fix thin, missing, or inconsistent content. Agents compare products using the information they can access, so unclear titles, incomplete descriptions, and uneven attributes can become a conversion problem on any platform, including Shopify.

WriteText.ai can help produce titles, descriptions, metadata, and alt text across a large catalog. It does not replace authoritative data for prices, stock, shipping, returns, or other store policies, which must continue to come from the systems responsible for them.

It is one thing to confirm the connection is set up correctly, and it is another to confirm a customer can complete a purchase during a busy sales period. You get proof of both: conformance results showing the connection meets the UCP standard, and full tests that runs real purchases from start to finish.

The tests cover the situations where these connections usually break: stock running out between the moment an agent finds a product and the moment it pays, a login expiring mid-purchase, an order confirmation arriving twice or out of order, a declined card, and a timeout partway through. This is part of our wider approach to automated testing for critical journeys.

1. Start with a fixed-price technical audit. We begin with a technical audit of your ecommerce stack. The price is fixed and agreed before work begins. The resulting capability matrix shows which UCP requirements are available now, available with configuration, require development, or are blocked by a system limitation.
2. Move from findings to implementation. Project Scoping turns the audit findings into an agreed scope and price for a Fixed Price Project. If your team wants to shape the solution hands-on with our developers, Co-Dev is the alternative.
3. Keep the specification current. After launch, Continuous Monthly Development can cover UCP specification upkeep, including assessing new dated releases and updating the implementation when needed.