News
CorgiMaps adds support for A2A, the Agent2Agent protocol
Brands can now turn on A2A, so other AI agents can search their catalog, build a cart and check out with them in a single conversation.

ATLANTA, July 27, 2026 – CorgiMaps today added support for A2A, the Agent2Agent protocol, alongside UCP, ACP, ONX and MCP. With A2A turned on, a brand automatically gets its own agent that other AI agents can find and work with directly, from browsing the catalog through checkout and order tracking.
A different flavor of agentic commerce
Most of the attention in agentic commerce has gone to checkout protocols. UCP and ACP define how a shopping assistant reads a store's catalog and completes a purchase for the shopper.
A2A solves a different problem. It's an open standard that lets AI agents built by different companies, on different frameworks, talk to each other, hand off tasks and return results. Google created A2A in 2025 and contributed it to the Linux Foundation that June, and more than 150 organizations were supporting it by August. Its steering committee includes Google, Microsoft, AWS, Salesforce, SAP, ServiceNow, Cisco and IBM.
That difference matters for commerce. A checkout protocol treats the store as a set of tools for the shopper's agent to call. With A2A, the store is an agent in its own right. The two agents can negotiate, ask clarifying questions and carry a task across a conversation. A2A's own documentation uses ordering as its example: agents "can negotiate or ask for clarification when placing an order."
Agents working with people are a growing area for commerce. Today's assistants mostly help one person shop in one session. Increasingly, agents will act for people and businesses over longer stretches of time, and they'll need to coordinate with other agents to get the job done.
Built for B2B as much as B2C
A2A is as much a B2B protocol as a consumer one. Agent-to-agent work fits business buying naturally: a retailer's purchasing agent checking a supplier's catalog, a distributor confirming availability, a wholesale buyer reordering stock. Tyson Foods and Gordon Food Service were early examples, building collaborative A2A systems to drive sales and reduce friction in their supply chain.
For brands that sell direct to consumers and through wholesale or retail partners, that makes A2A relevant to both sides of the business.
What brands get
With CorgiMaps A2A support, a brand gets a published agent card that describes what its agent can do, plus fully functional agentic endpoints allowing connecting agents to:
- Search the catalog and look up products
- Build a cart and complete a checkout
- Check order and fulfillment status
It all happens in one conversation, on the same product, pricing and inventory data the brand already uses for UCP, ACP and ONX. A2A is also advertised as a transport for UCP checkout, so a UCP agent can complete a purchase over A2A.
Brands can test every A2A action from CorgiMaps protocol testing experience in chat before going live.
"Checkout protocols were the first wave, and they matter. But a lot of commerce is going to be agents working with other agents, on behalf of people and businesses," said Jonathan Porter, co-founder and Head of Product & Engineering at CorgiMaps. "Nobody knows yet which approach wins, so we want our brands present in all of them. A2A gives a brand's store an agent of its own, running on the same data it already trusts everywhere else."
A2A support is available today to all CorgiMaps customers. Simply turn it on under Protocol Settings, no new data required.
About CorgiMaps
CorgiMaps connects a brand's existing commerce and supply chain systems to the agentic commerce protocols that AI shopping agents use, including UCP, ACP, ONX, A2A and MCP. Brands connect their data and CorgiMaps publishes protocol-compliant data structures and endpoints so AI platforms can find, recommend and sell their products. Orders flow back into the brand's existing systems with no changes to fulfillment. Learn more at corgimaps.com.