Google’s Agent2Agent (A2A) protocol has become a hosted project of the Agentic AI Foundation (AAIF), the Linux Foundation body that already stewards Anthropic’s Model Context Protocol (MCP), according to an AAIF announcement published August 17.

The move places the two most widely adopted agent-interoperability standards under one neutral governance structure for the first time. Google launched A2A in April 2025 and later donated it to the Linux Foundation; the protocol reached a stable 1.0 specification in March 2026, adding multi-tenancy support, version negotiation and cryptographically signed “agent cards” that let one agent verify another’s identity before handing off work to it.

AAIF says more than 150 organizations now back A2A, including founding supporters AWS, Cisco, Google, Microsoft, Salesforce, SAP and ServiceNow. Cloud implementations already ship inside Google Cloud’s Agent Engine, Microsoft’s Azure AI Foundry and AWS Bedrock AgentCore, and the foundation cites production use inside Huawei’s Celia assistant and Tencent’s WeChat.

Two layers of one stack

AAIF frames A2A and MCP as complementary rather than competing. MCP standardizes how a single agent reaches a tool, database or file system; A2A standardizes how one agent discovers and delegates a task to another agent, including one built by a different vendor on a different framework. “A2A represents an important step toward an open, interoperable future” for agents that need to work across vendor lines, AAIF chief technology officer Manik Surtani said in the announcement. Google Cloud vice president Rao Surapaneni added that A2A “enables agents to easily discover, delegate, and collaborate across frameworks” without custom point-to-point integrations.

AAIF itself launched in December 2025, co-founded by Anthropic, Block and OpenAI with backing from Google, Microsoft, AWS, Cloudflare and Bloomberg. It has since grown from 49 founding members to more than 250, spanning AI labs, cloud providers and enterprise software vendors that now track agent-interoperability standards through a single body rather than separate, vendor-run efforts.

For developers, the consolidation means fewer places to check for updates on how autonomous agents built with different tools can find and hand off work to each other — a question that has grown more pressing as companies move from single chatbots to fleets of agents running in production. A companion explainer on how A2A differs from MCP breaks down the technical distinction in more depth.