About the A2A Community: Tutorials, Documentation, and Discussion

About the A2A Community: Tutorials, Documentation, and Discussion

What this site is

This A2A Protocol documentation and developer community shares tutorials, technical topics, development guides, and resources to help you build agents that work together:

  • explainers for the protocol specification and its data structures;
  • topics on core concepts, the task lifecycle, streaming, and agent discovery;
  • a tutorial for building an A2A agent from scratch;
  • implementation guides covering integration choices, troubleshooting, recovery, and interoperability testing;
  • version changes, SDKs, partners, and community news.

Start with a tutorial, explore collaboration patterns in the technical topics, and find SDKs, tools, and protocol documentation. Pages retain source and version links for further reading and comparison.

A2A Protocol at a glance

The Agent2Agent (A2A) Protocol is an open standard for communication and interoperability between opaque agentic applications. No matter which vendor built an agent or which framework it uses, A2A lets agents discover each other’s capabilities, negotiate how to interact, and collaborate securely on tasks, without exposing their internal memory, tools, or proprietary logic.

A2A was originally developed by Google and donated to the Linux Foundation. It is now a Growth Stage project at the Agentic AI Foundation (AAIF), maintained by a Technical Steering Committee (TSC), and licensed under the Apache License 2.0. A2A Protocol v1.0 was released in March 2026, followed by the v1.0.1 bug-fix release.

A2A complements the Model Context Protocol (MCP): MCP connects agents to tools and data, while A2A connects agents to other agents. See A2A and MCP .

Core building blocks

  • Agent Card: JSON metadata an agent publishes to describe its identity, capabilities, skills, supported interfaces, and authentication, so clients can discover and select it.
  • Task: a stateful process in which a client and a remote agent collaborate toward a goal, with a well-defined lifecycle.
  • Message / Part / Artifact: messages carry interaction content, a Part is the smallest unit of content, and an Artifact is a result produced by a task.
  • Protocol bindings: one data model, available through the JSON-RPC, gRPC, and HTTP+JSON/REST bindings.
  • Security and extensions: enterprise-grade authentication and authorization, Agent Card signatures, multi-tenancy, and an extension mechanism for adding new capabilities.

These concepts are covered in Key Concepts and the Specification .

Where to start

Join the community

Share agent collaboration experience, propose features, and discuss problems through these A2A project channels:

The official repository is a2aproject/A2A . More entry points are in the Community Hub .

Contact

For content suggestions, corrections, or collaboration, email ai@agent2agent.info .