Artificial Intelligence 8 min read

Google Unveils the Open‑Source Agent2Agent (A2A) Protocol and Highlights Its Enterprise Adoption

At Google Cloud Next 25, Google introduced the open‑source Agent2Agent (A2A) protocol—a standardized interaction model for AI agents that breaks system silos, supports major enterprise platforms, follows five design principles, and is already being adopted by dozens of leading companies across various industries.

DevOps
DevOps
DevOps
Google Unveils the Open‑Source Agent2Agent (A2A) Protocol and Highlights Its Enterprise Adoption

On April 9, Google announced at Google Cloud Next 25 the open‑source Agent2Agent (A2A) protocol, the first standard for intelligent agent interaction.

A2A aims to eliminate system islands, enhancing agents' capabilities, cross‑platform interoperability, and execution efficiency, with early support from major enterprise applications such as Atlassian, Box, Cohere, Intuit, Langchain, MongoDB, PayPal, Salesforce, SAP, ServiceNow, UKG, and Workday.

The protocol is likened to the early Android ecosystem, with over 50 prominent enterprises joining, which is expected to boost A2A's commercial value and accelerate the broader agent ecosystem.

Google also released an Agent Development Kit (ADK), internal testing tools, an Agent Engine, and a new Agent Marketplace, mirroring OpenAI's recent open‑source efforts.

A2A is an open protocol that provides a standard way for agents to collaborate regardless of underlying frameworks or vendors, enabling seamless communication between disparate enterprise platforms.

Examples illustrate how a large e‑commerce company can integrate agents from Atlassian, Box, Salesforce, and Workday using A2A to automate data exchange without rebuilding complex data pipelines.

Google follows five key principles when designing the protocol: focusing on natural, unstructured agent collaboration; building on existing standards like HTTP, SSE, and JSON‑RPC; supporting enterprise‑grade authentication and authorization compatible with OpenAPI; offering flexibility for tasks ranging from quick actions to multi‑day research; and supporting multimodal content such as audio, images, and video streams.

The protocol defines "Agent Cards" in JSON that let client agents discover the most suitable remote agents for a given task, after which tasks are assigned via A2A communication.

Task management is central to A2A: client and remote agents exchange messages, artifacts, and status updates, with completed outputs referred to as "artifacts".

A2A also negotiates user‑experience aspects, allowing agents to agree on content types (e.g., iframes, video, web forms) to deliver optimal experiences based on device capabilities.

Since its launch, more than 50 well‑known enterprises—including Accenture, Boston Consulting Group, Capgemini, Deloitte, Oracle, HCL Technologies, Infosys, KPMG, SAP, McKinsey, PwC, and others—have joined the A2A initiative.

Technical companies such as Atlassian (Jira, Confluence), Box, Intuit (QuickBooks, TurboTax), and MongoDB have also integrated A2A, enabling their services to work seamlessly with agents for improved workflow automation and data management.

Overall, Google seeks to unify fragmented agent interactions under a common standard, offering a more powerful execution and interaction model than previous efforts like MCP.

AI agentsOpen-sourceA2AEnterprise Integrationgoogle cloudAgent2Agent
DevOps
Written by

DevOps

Share premium content and events on trends, applications, and practices in development efficiency, AI and related technologies. The IDCF International DevOps Coach Federation trains end‑to‑end development‑efficiency talent, linking high‑performance organizations and individuals to achieve excellence.

0 followers
Reader feedback

How this landed with the community

login Sign in to like

Rate this article

Was this worth your time?

Sign in to rate
Discussion

0 Comments

Thoughtful readers leave field notes, pushback, and hard-won operational detail here.