A2A is the open standard for inter-agent communication and is how agents discover each other, delegate tasks, and work across any framework or vendor.
| TL;DR |
|---|
| Agent2Agent (A2A), the open standard for inter-agent communication, is joining the Agentic AI Foundation (AAIF) as a hosted project. A2A defines how agents discover each other, delegate tasks, and exchange work, regardless of what framework they were built in or what company made them. It's backed by over 150 organizations and running in production across supply chain, financial services, and mobile platforms at OS scale. AAIF is the open, neutral home where the agentic stack is being built. |
When engineers in supply chain and financial services started deploying multi-agent systems in production, they encountered the same problem repeatedly. Agents built on different frameworks couldn't hand off work to each other without custom integration code written specifically for that pairing. Every new vendor relationship required the same integration work from scratch. The cost wasn't in the agents themselves but in the integrations between them.
A2A was built to solve that problem directly.
How A2A works in practice
A2A is an open standard that defines how agents communicate across framework and vendor boundaries. An agent publishes an agent card containing a structured description of what it can do and how to reach it. Other agents read that card, discover capabilities, and delegate tasks without a human brokering the handoff. The exchange is structured, observable, and framework agnostic.
Google launched A2A in April 2025 and donated it to the Linux Foundation with founding organizations like AWS, Cisco, Google, Microsoft, Salesforce, SAP, and ServiceNow. In August 2025, IBM's Agent Communication Protocol merged into A2A showing that the field was moving toward a single shared standard rather than competing ones.
A2A v1.0, the first stable specification, shipped in March 2026. It added multi-protocol bindings and version negotiation, multi-tenancy, and signed agent cards for cryptographic identity verification.
Where A2A is already running
A2A is already running in production across mobile platforms, cloud AI infrastructure, financial services, supply chain, and enterprise IT.
Huawei has standardized A2A as the protocol between Celia, its OS-level AI assistant, and in-app agents across the HarmonyOS developer platform. The supported scenarios include Celia handing long running tasks to an app agent, Celia controlling app UI through an agent, and Celia requesting contextual recommendations from an application agent. This is agent communication between an OS assistant and an application running at mobile platform scale.
Tencent's WeChat is among the first major apps integrating with Huawei and other Android OEM assistants over A2A, enabling messages, voice calls, and video calls initiated through an AI assistant. The flow runs with dual authorization through the A2A protocol.
Beyond mobile platforms, A2A is built into the leading cloud AI platforms. Google Cloud supports developing and deploying A2A agents through ADK, Agent Engine, Cloud Run, and GKE. Microsoft Azure AI Foundry lets agents expose A2A endpoints and find external agents through standard discovery. AWS Bedrock AgentCore can host and operate A2A servers, so agents built with different frameworks or hosted in different clouds can communicate through the same protocol.
Google Cloud and PayPal are extending A2A into agentic commerce through the Agent Payments Protocol (AP2). Shopping and merchant agents communicate over A2A throughout product discovery, pricing, and order fulfillment, with AP2 providing the payment authorization layer.
Where A2A fits in the agentic stack
AAIF‘s projects each cover distinct layers of the open agentic stack.
Instructions and context: AGENTS.md standardizes how projects communicate expectations, conventions, and operating instructions to AI agents.
Agent runtime: goose provides the environment in which an agent reasons, plans, invokes capabilities, and carries out work.
Agent to tool connectivity: MCP standardizes how agents connect to and interact with tools, data sources, applications, and services.
Traffic mediation and control: agentgateway handles operations. It sits at the boundary between agent systems and the infrastructure they run on, managing routing, policy, and observability.
Agent to agent interoperability: A2A adds an agent to agent interoperability layer by standardizing how independent agents discover one another, communicate, delegate tasks, and exchange results across systems and organizational boundaries.
“A2A represents an important step toward an open, interoperable future for AI agents,” said AAIF CTO Manik Surtani. “By bringing the project to the Agentic AI Foundation, we're creating a neutral home where the community can collaborate on the standards that will enable agents from different vendors, frameworks, and organizations to work together. We're excited to welcome the A2A community and build this future together.”
“As organizations move multi-platform agentic systems into production, interoperability is becoming essential,” said Rao Surapaneni, VP & General Manager, Business Application Platforms, Google Cloud. “A2A enables agents to easily discover, delegate, and collaborate across frameworks and siloed platforms, without custom integrations. Bringing A2A to the Agentic AI Foundation further empowers enterprises to build and scale agentic systems on a truly open foundation.”
Why neutral governance matters for infrastructure
The Linux Foundation model for open standards has a track record. When foundational components are owned by a single vendor, every downstream team absorbs that vendor's roadmap constraints and release cycles. When they're governed openly, the community shapes what gets built and when.
A2A's 150+ partner organizations include direct competitors. That breadth only holds together under governance that no single participant controls. The Agentic AI Foundation, hosted by the Linux Foundation, provides that structure. It also means the full stack, from context to communication to operations to the open agent layer, is governed in the same way and in the same place.
For teams building multi-agent systems, this removes a specific category of risk. The protocol your agents depend on for cross-framework communication isn't subject to a single vendor's product decisions.
We're glad A2A is here. Come build with us.
- Explore the A2A spec and SDKs: https://a2a-protocol.org/latest/
- Join the conversation on Discord: https://discord.com/invite/a2aprotocol
- Tune in to A2A’s Technical Steering Committee Meetings: https://zoom-lfx.platform.linuxfoundation.org/meetings/agent2agent
Share
Author
