White Paper
Navigating the Agentic Commerce Era On Your Terms
The practical guide to help you approach agentic commerce strategically — not out of FOMO.
Introduction: Understanding agentic protocols
Agentic commerce offers brands and retailers an array of opportunities, but it’s not without challenges. One of them is understanding and implementing multiple protocols that fulfill specific roles, such as MCP, ACP and UCP.
Let’s break them down in a way that makes sense and see why retailers need to abstract all protocols into a single commerce layer, so they can focus on strategy, growth and customer experience.
MCP (Model Context Protocol): Giving AI agents context
MCP is a protocol created by Anthropic that standardizes how AI agents securely connect to external data sources, tools and APIs. While it can technically be used by individual AI models, such as ChatGPT, Gemini and Claude, its strongest use case is enabling autonomous agents to dynamically discover, negotiate and interact with unfamiliar systems through a structured process.
This makes MCP particularly powerful in scenarios where an AI agent must independently determine how to authenticate, interpret capabilities and execute actions across heterogeneous environments.
This means AI agents can access live, contextual information beyond their training data while respecting privacy and permission boundaries. In practical terms, MCP acts as a bridge between agent models and the real-world systems they interact with, enabling them to access commerce data, execute tasks and retrieve up-to-date information safely and consistently.
What MCP solves:
- AI system integration with secure context, tool access and real-time data for agents, e.g., an agent can remember that a shopper prefers eco-friendly brands, check inventory levels or loyalty program eligibility before making recommendations.
- Autonomous shopping agents that buy on behalf of customers when MCP is connected to a commerce engine.
What merchants care about:
- A smart agent increases relevance, improves customer experience and drives more informed purchase decisions.
- AI agents can manage the customer journey end-to-end while your business remains in control as the merchant of record.
ACP (Agentic Commerce Protocol): Frictionless checkout in AI ecosystems
ACP, created by OpenAI and Stripe (with support from commercetools), focuses on secure, instant checkout, primarily for single-item purchases in GenAI environments like ChatGPT and Microsoft Copilot.
In March 2026, OpenAI paused Instant Checkout in ChatGPT. ChatGPT now focuses on product discovery, with checkout completed on the merchant’s site. OpenAI is now testing product checkout plugins in ChatGPT, currently in beta with approved partners.
What ACP solves:
- A shared standard for agentic checkout, using delegated payment models via Stripe.
- Connects AI platforms to the merchant’s checkout, so a purchase that starts in the AI interface can be completed.
- Most relevant for the OpenAI ecosystem, where the checkout approach is still evolving.
Scenario: A shopper finds a book recommended by ChatGPT and taps through to the retailer’s site, where the purchase completes at the retailer’s checkout. As OpenAI’s checkout plugins mature, more of this journey may happen inside ChatGPT.
UCP (Universal Commerce Protocol): Making products discoverable and shoppable anywhere
UCP, developed by Google and partners including Walmart and Target, standardizes the entire shopping lifecycle for AI agents, from discovery and comparison to checkout, order tracking and post-purchase support.
As the name suggests, this protocol provides a “universal” language that standardizes how AI agents interact with merchant systems. Merchants can publish their capabilities and product details once, and any UCP-compliant agent can use them.
What UCP solves:
- Publish your products and capabilities once, and any UCP-compliant AI agent can access them.
- UCP-compliant AI agent can manage carts, check out securely and even handle returns or refunds.
- Powers Google’s Universal Cart, live since 15 October 2026 for US shoppers in Search, AI Mode and Gemini.
- Designed to work with MCP and other Google-built protocols, including A2A (agent-to-agent protocol) and AP2 (agent payments protocol).
- Broader focus than ACP, it’s designed for long-term multi-agent interoperability.
Why merchants care about:
- Your products become discoverable across multiple AI platforms without multiple integrations.
- AI agents can manage the customer journey end-to-end.
- Through a modular design, UCP is built to support multiple payment providers. With Google’s native checkout today, payments are processed through Google Pay.
Scenario: A shopper asks Google’s AI agent to find a laptop under $1,000. UCP ensures your inventory and pricing appear accurately, lets the agent add the product to a cart and lets the shopper check out either on your site (Universal Cart transfer) or inside Google with Google Pay (native checkout)
What agentic protocol should you focus on?
Whether your business uses only one or multiple protocols for agentic commerce depends on your goals, use cases and commerce workflows. Consider these three scenarios:
MCP lets agents retrieve information from systems, maintain context across tasks and support analysis without touching transactions.
Agents can manage carts and complete checkout through MCP when it’s connected to your commerce engine, while you stay in control as the merchant.
You may need all three. MCP gives broad agent compatibility, with transactions running through your commerce engine. UCP covers Google’s AI surfaces, including Universal Cart. For the OpenAI ecosystem, ACP is the reference, though its checkout approach is still evolving
Addressing emerging agentic protocols with commercetools
For most retailers, it’s simply not feasible to build, maintain and update integrations for every emerging agentic commerce protocol on their own. Each new use case you want to implement and its corresponding standard — UCP, ACP, MCP and others — requires significant engineering effort, ongoing updates and specialized expertise.
Withcommercetools Agentic Commerce, merchants can sell through AI channels like Google’s AI Mode and Gemini. With commercetools MCPs, they can also power their own AI shopping experiences. Either way, we keep up with the protocols as they evolve.
By leveraging commercetools, retailers can stay ahead of emerging protocols without the heavy lift, enabling agentic commerce experiences across any channel.
FAQs
MCP, ACP and UCP serve different purposes in agentic commerce. MCP (Model Context Protocol) enables AI agents to securely connect to external systems, access data and use tools. ACP (Agentic Commerce Protocol) focuses on enabling secure checkout within AI ecosystems. UCP (Universal Commerce Protocol) standardizes the broader shopping journey, from product discovery and cart management to checkout and post-purchase interactions.
Not necessarily. The protocols a merchant needs depend on its business goals, target AI channels and desired customer experiences. MCP is useful for connecting AI agents to commerce systems and powering custom AI experiences. UCP supports shopping across Google’s AI surfaces and other compatible agents, while ACP is relevant to the evolving OpenAI checkout ecosystem. Some merchants may benefit from supporting all three.
With commercetools Agentic Commerce, merchants can support the protocols that matter without building each integration themselves.
Yes, in some channels. On Google’s AI surfaces, shoppers can check out through Universal Cart, either on the merchant’s site or inside Google using Google Pay. In ChatGPT, checkout currently completes on the merchant’s site while OpenAI tests checkout plugins.
Yes. Although OpenAI paused Instant Checkout in ChatGPT in March 2026, ACP remains relevant to the development of agentic checkout. ChatGPT currently focuses on product discovery, with purchases completed on merchant websites, while OpenAI tests product checkout plugins with approved partners. The checkout experience is evolving, making it important for merchants to monitor developments in the OpenAI ecosystem.
UCP is an open standard developed by Google and industry partners to enable AI agents, businesses and payment providers to communicate throughout the shopping journey. It supports capabilities such as product discovery, cart management, checkout and post-purchase interactions. UCP powers Google’s Universal Cart, which allows shoppers to build baskets across Google’s AI shopping experiences, including Search, AI Mode and Gemini.
Retailers can use a commerce platform that abstracts protocol complexity and connects AI channels to existing commerce capabilities. commercetools Agentic Commerce supports connections to emerging AI shopping channels, including Google's AI experiences, while commercetools MCPs help merchants power their own AI shopping experiences. This reduces the need to build and maintain integrations for each protocol independently.
These protocols enable AI agents to interact with existing commerce infrastructure rather than requiring retailers to replace their commerce platforms. They connect AI experiences to capabilities such as product data, pricing, inventory, carts, checkout and order management. With commercetools, merchants can extend their existing commerce capabilities to AI channels while keeping their core commerce operations connected.
Agentic commerce protocols are continuing to evolve, with different standards addressing different parts of the shopping journey. Retailers should prioritize flexible commerce infrastructure that can support new protocols and AI channels as they emerge. Rather than building and maintaining every integration independently, merchants can use a composable commerce platform to adapt their capabilities as the agentic commerce ecosystem develops.
Protocols enable merchants to make their products discoverable across multiple AI platforms without building a separate integration for each platform. AI agents can support the shopping journey from discovery to checkout while the merchant remains in control. A platform that supports protocols as they evolve also reduces the engineering effort required to build and maintain each one.
