AI Search & Agents
Universal Commerce Protocol (UCP)
In shortAn open standard led by Google for AI agents to discover products, build carts and check out with merchants; announced January 2026.
The Universal Commerce Protocol (UCP) is an open-source standard that gives AI agents, shopping surfaces, merchants and payment providers a common language for the whole buying journey, from product discovery to checkout and order management. Google announced it with Shopify at the National Retail Federation show in January 2026, with Etsy, Wayfair, Target and Walmart as co-developers and more than 20 further partners, including payment networks, at launch.
How it works
UCP defines standard building blocks (“capabilities”) that a business can expose and an agent can use. According to the protocol site, these include:
- Catalogue: searching and retrieving products and prices.
- Cart and checkout: building a cart and completing a purchase, including tax and pricing logic.
- Identity linking: connecting a shopper’s merchant account through OAuth 2.0 without sharing passwords.
- Order management: status, tracking and returns through webhooks.
UCP is designed to work with other agent standards: the Agent Payments Protocol (AP2) for payments, Agent2Agent (A2A), and the Model Context Protocol. Google uses it to power checkout inside AI Mode and Gemini.
In MLforSEO’s AI Search Optimisation & Agentic SEO course, UCP sits in the “action layer” alongside WebMCP and OpenAI and Stripe’s ACP: ACP is the fast lane for buying inside a chat, while UCP handles complex, multi-step retail flows such as identity verification, inventory sync and enterprise system integration.
How it’s used in SEO and marketing
- Making products purchasable by agents in Google’s AI surfaces without the shopper visiting your site.
- Raising the bar for product data quality: prices, stock, shipping and return policies must be accurate and machine-readable.
- Prompting ecommerce teams to align their Merchant Center feeds, schema and platform integrations.
Example
A shopper asks AI Mode for a waterproof hiking jacket under £150 that can arrive by Friday. An agent searches catalogues, filters by price, size and delivery date, and offers checkout in the conversation. A merchant whose stock and delivery promises are exposed in structured, verifiable form can be chosen; one whose return policy only exists in prose may be skipped.
Limitations and things to watch
- UCP is new and evolving. Check the specification and Google’s merchant documentation for current requirements and supported regions.
- Implementation usually comes through your commerce platform; few businesses will build it from scratch.
- The course advice is not to rush: the preconditions (clean entity data, complete product schema, typed policy fields, API-ready inventory) matter now and help with every protocol.
Learn it on MLforSEO
Where Universal Commerce Protocol (UCP) shows up on MLforSEO
Learn it in a course
Sources and further reading
Other terms from AI Search & Agents
- Agentic AIAI systems that pursue a goal over several steps on their own: planning, choosing tools or sources, acting and…
- Agentic Commerce Protocol (ACP)An open standard from OpenAI and Stripe defining how AI agents complete purchases with businesses on a buyer's…
- Agentic SearchSearch where an AI system plans, runs and evaluates many searches itself, then delivers one synthesised answer…
- AI AgentA system that pursues a goal on its own: it plans, uses tools, keeps track of what it has learned and adapts u…
- AI Citation (LLM Citation)A source link or named mention an AI search system attaches to its answer, showing which pages or brands it dr…
- AI Crawlers (GPTBot, ClaudeBot, PerplexityBot)Bots run by AI companies that fetch web pages for model training, AI search indexes or live answers; you contr…
- AI Mode (Google)Google AI Mode is a conversational search experience that answers with a Gemini-written response, built by fan…
- AI OverviewGoogle's AI-generated answer shown above some search results, summarising information from several pages and l…
- AI Rank Tracking (Prompt Tracking)Monitoring whether and how AI assistants mention or cite your brand by running a fixed set of prompts on a sch…
