Agent2Agent (A2A) Protocol
An open protocol, backed by the Linux Foundation, that lets AI agents built on different frameworks discover each other, negotiate how to interact, and collaborate securely on long-running tasks without exposing their internal state.
Agent2Agent (A2A) is an open protocol for communication and interoperability between AI agents that would otherwise be opaque to each other, built on different frameworks by different organizations and running on separate servers. It gives agents a common language so they can discover each other's capabilities, negotiate interaction modes such as text, forms or media, and collaborate securely on long-running tasks, without needing to share their internal state, memory or tools.
It complements the Model Context Protocol, which connects an agent to tools, by letting agents collaborate directly with each other instead. A companion course, built with Google Cloud and IBM Research, covers making agents A2A-compliant from frameworks such as Google ADK, LangGraph or BeeAI, creating A2A clients, orchestrating sequential and hierarchical agent workflows, and building a multi-agent system.
It is licensed Apache-2.0 and is a Linux Foundation project.