Google Announces the Agent2Agent Protocol
On April 9 2025 Google announced the Agent2Agent protocol (A2A) with more than 50 partners. We examine what A2A specifies — Agent Cards, task lifecycles, artifacts — how it complements Anthropic's MCP, what the draft deliberately leaves open, and why governance will decide whether it becomes the interoperability layer for multi-agent systems.
Agents That Cannot Talk to Each Other
Enterprises are beginning to run more than one AI agent. A procurement agent built on LangGraph, a support agent on CrewAI, a vendor agent inside Salesforce or SAP. Each framework defines its own message formats, its own state model, its own calling conventions. Two agents from different stacks cannot delegate work to each other without custom glue code. Every new pairing is a bespoke integration, and the integration count grows quadratically with the number of agents.
One half of the problem already has a standard. Since November 2024, Anthropic's Model Context Protocol (MCP) defines how an agent connects to tools and data. The other half — how one agent talks to another agent — had none. Point-to-point bridges exist, but none of them is a standard. That is the gap Google now addresses.
What Google Announced on April 9
On April 9, 2025, at Cloud Next in Las Vegas, Google announced the Agent2Agent protocol (A2A): an open protocol for communication between AI agents, launched with support from more than 50 technology partners — among them Atlassian, Box, Cohere, LangChain, MongoDB, PayPal, Salesforce, SAP, ServiceNow, and Workday — plus service providers including Accenture, Deloitte, and McKinsey.
The specification is a draft, published as open source on GitHub, announced alongside Google's open-source Agent Development Kit (ADK) and an AI Agent Marketplace. A2A builds on HTTP, Server-Sent Events, and JSON-RPC — deliberately unspectacular standards that pass through enterprise firewalls. The model defines two roles: a client agent formulates tasks; a remote agent executes them. Google states that a production-ready version is planned for later in 2025.
Agent Cards and Capability Discovery
Discovery works through the Agent Card: a JSON document that a remote agent serves at a well-known URL, typically /.well-known/agent.json. The card declares identity, skills, endpoint, supported input and output modalities, and authentication requirements. A client agent reads the card and decides whether this counterpart can handle a given task — machine-readable capability advertisement, comparable to what OpenAPI descriptions did for REST APIs.
The card is the whole discovery story so far. The draft defines no registry, no search infrastructure, no vetting mechanism. How a client finds candidate agents in the first place, and why it should trust them, remains outside the specification.
Tasks as the Unit of Work
A2A communication is oriented around tasks. A task is a stateful object with a unique ID and a defined lifecycle: submitted, working, input-required, completed, failed, canceled. Outputs are artifacts, composed of parts with negotiated content types — text, files, structured data. The protocol is modality-agnostic by design; audio and video are named as goals.
Long-running work is a core case, not an edge case. Status updates stream over SSE, push notifications cover tasks that run for hours or days, and the input-required state models human-in-the-loop steps. This is the central design decision: A2A treats inter-agent work as asynchronous jobs with observable state, not as a chat transcript.
A2A and MCP Are Complements
Google positions A2A explicitly as a complement to Anthropic's MCP, announced on November 25, 2024. Amin Vahdat, Google Cloud's VP for ML, Systems and Cloud AI, summarized the split at launch: MCP connects models to data and tools; A2A provides interoperability between agents. In practice an agent will use both — MCP downward to its tools, A2A sideways to its peers.
The distinction is technically meaningful. An MCP tool is called; it has no opinions. An A2A remote agent is delegated to; it plans, asks follow-up questions, and may refuse. The two protocols encode exactly that difference: function-call semantics on one side, task lifecycles on the other.
| Aspect | MCP (Anthropic) | A2A (Google) |
|---|---|---|
| Announced | November 25, 2024 | April 9, 2025 |
| Connects | Agent to tools and data | Agent to agent |
| Interface unit | Tools, resources, prompts | Tasks with artifacts |
| Counterpart | Deterministic tool call | Opaque autonomous agent |
| Discovery | Configured servers | Agent Card at a well-known URL |
What A2A Does Not Do
A2A does not share memory, context windows, or tools between agents. Agents remain deliberately opaque to each other; only tasks, messages, and artifacts cross the boundary. A2A also does not orchestrate: which agent receives which task, and in what order, remains the application's decision.
Nor does the protocol guarantee semantic interoperability — two agents can both speak A2A and still misinterpret each other's artifacts. And the specification is a draft. Authentication follows OpenAPI practice, but cross-organization identity, agent registries, and liability for delegated actions are open questions. Production use in April 2025 means building on a moving target. We treat A2A accordingly: as an interface to prototype against, not yet a foundation to depend on.
Open Governance as a Signal
The partner list is itself a governance statement. Salesforce, SAP, ServiceNow, and Workday compete with Google in several markets; their names at launch signal that A2A is meant as an industry protocol, not a Google API. Google says it will develop the protocol in the open with clear contribution pathways, and has released the draft under an open-source license.
History teaches caution and hope in equal measure. Protocols gain value through neutrality; standards controlled by a single vendor tend to fragment. The credible next step would be transferring A2A to a neutral foundation with shared governance. Whether Google takes that step will say more than any detail of the specification.
The Outlook From April 2025
From where we stand in April 2025, the likely shape of the stack is two thin layers: MCP for agent-to-tool, A2A for agent-to-agent — provided the production-ready version lands this year and independent implementations follow. The open questions we are watching: registry and discovery infrastructure, verifiable agent identity, and whether Anthropic and OpenAI adopt A2A or answer with protocols of their own.
Our conclusion is practical. Design agent boundaries as task interfaces today — explicit inputs, observable state, typed artifacts — because that structure maps onto A2A whatever the final specification looks like. Within twelve months, A2A either becomes the default interoperability layer for enterprise agents or splinters into vendor dialects. Fifty partners at launch argue for the former.
Sources
- Announcing the Agent2Agent Protocol (A2A) — Google Developers Blog (April 9, 2025)
- Build and manage multi-system agents with Vertex AI — Google Cloud Blog (April 9, 2025)
- Google's Agent2Agent Protocol Helps AI Agents Talk to Each Other — The New Stack (April 9, 2025)
- Introducing the Model Context Protocol — Anthropic (November 25, 2024)
- A2A protocol draft specification — GitHub google/A2A (April 2025)