News · April 9, 2025
Google proposes an open standard for AI agents to talk to each other
Google published the Agent2Agent protocol, an open specification for agents from different vendors to exchange tasks and results. It sits next to MCP and points to a future where business software is wired together by agents.
What happened
- Google announced the Agent2Agent protocol (A2A) on April 9, 2025 as an open protocol for AI agents to communicate and coordinate across platforms built by different vendors, according to the Google Developers Blog.
- Google said A2A launched with more than 50 technology and services partners and is built on existing standards such as HTTP, server-sent events and JSON-RPC, with agents describing their abilities in a JSON Agent Card.
- The company positioned A2A as complementary to Anthropic's Model Context Protocol and described the current specification as a draft, with a production-ready version planned later in the year.
- Virtualization Review reported that A2A was unveiled at Google Cloud Next '25 and compared MCP to a plugin layer for a single agent and A2A to a networking layer between agents.
Why it matters
Standards matter more than any single model release, because they decide how much of today's integration work survives the next platform change. Teams building internal agents should keep them behind clean, documented interfaces so they can be exposed over MCP, A2A or whatever wins without a rewrite.
It is still a draft. We would not bet a production roadmap on it yet, but a small proof of concept between two internal services is a cheap way to learn what agent-to-agent traffic means for authentication, logging and cost control.
Sources
- Google Developers Blog: Announcing the Agent2Agent Protocol (A2A)
- Virtualization Review: Protocols for Agentic AI: Google's New A2A Joins Viral MCP
This post is CMM’s own summary and commentary on publicly reported news. Facts are attributed to the sources linked above; please read them for full details. CMM is not affiliated with these publishers or with the companies mentioned, and trademarks belong to their owners. Prepared with AI assistance. Corrections or requests: [email protected]. See our editorial policy.
Want help with this?
Tell us what you need, and we will suggest the right team model on a short call.
Talk to us