Agent-to-Agent (A2A) is a protocol developed by Google that allows AI agents from different vendors to communicate and collaborate with each other.
Agent-to-Agent (A2A) is a protocol developed by Google that allows AI agents from different vendors to communicate and collaborate with each other. While MCP connects models to tools, A2A connects agents to other agents, enabling multi-agent workflows where specialized agents hand off tasks to each other.
A2A solves the problem of agent interoperability. Without it, an agent built with one framework can't easily talk to an agent built with another. A2A defines how agents discover each other's capabilities, negotiate tasks, and exchange information. It's designed to work alongside MCP — agents use MCP to access tools and A2A to coordinate with other agents.
A research agent finds relevant papers, then hands them to a summarization agent via A2A, which then passes the summary to a writing agent that drafts a report.
Browse our complete glossary or subscribe to our newsletter for the latest AI news and insights.