
Microsoft announces support of Google's new Agent-to-Agent protocol
Microsoft has announced its commitment to supporting Google’s newly introduced open protocol, known as Agent2Agent (A2A), which facilitates communication among AI "agents."
Microsoft has announced its commitment to supporting Google’s newly introduced open protocol, known as Agent2Agent (A2A), which facilitates communication among AI "agents."
This announcement, confirms that Microsoft will integrate A2A into two of its AI development platforms: Azure AI Foundry and Copilot Studio. Additionally, Microsoft has joined the A2A working group on GitHub to actively contribute to the development of the protocol and its associated tools. In its official blog post, Microsoft articulated, “By supporting A2A and building on our open orchestration platform, we are laying the foundation for the next generation of software—collaborative, observable, and adaptive by design.”
The company highlighted that the most effective AI agents will not be restricted to a single application or cloud environment; rather, they will function seamlessly across various applications, domains, and ecosystems. The A2A protocol, unveiled by Google in early April, allows AI agents—semi-autonomous programs equipped with artificial intelligence—to collaborate across multiple clouds, applications, and services. By employing this protocol, agents can exchange objectives and invoke actions, while offering developers a suite of interoperable components that ensure secure collaboration between agents.
Upon the implementation of A2A support within Azure AI Foundry and Copilot Studio, agents created using these platforms will be empowered to engage with external agents, including those developed through different tools or hosted outside of the Microsoft ecosystem. For example, a Microsoft agent may facilitate the scheduling of a meeting, while a Google agent composes the email invitations. In its blog post, Microsoft further explained, “[C]ustomers can build complex, multi-agent workflows that encompass internal agents, partner tools, and production infrastructure while upholding governance and service-level agreements.” This strategic decision aligns with the broader industry movement towards shared agent protocols. Microsoft’s endorsement of A2A follows its previous introduction of support for the MCP standard, formulated by Anthropic, which connects AI capabilities to localized data systems within Copilot Studio. Earlier this year, other prominent AI model providers, including Google and OpenAI, also announced their adoption of the MCP standard.