Google has just announced the Agent2Agent Protocol (A2A). A2A is open source and aims to enable AI agents to work together seamlessly, potentially multiplying productivity gains in end-to-end business processes.
As I understand it, A2A is to agent communication what MCP is to tool use. At the time, I saw MCP as an opportunity to reduce frictions in agent deployment while maintaining a level of security (see here), and it has taken off since then. Google’s A2A seems to take it to the next level, providing more security in the cloud for multiple agents to communicate and collaborate:
A2A focuses on enabling agents to collaborate in their natural, unstructured modalities, even when they don’t share memory, tools and context. We are enabling true multi-agent scenarios without limiting an agent to a “tool.”