Agent2Agent Protocol
- In Turkish
- Agent2Agent Protokolü
In short
The Agent2Agent (A2A) protocol is an open standard that lets AI agents built by different teams or vendors find each other, hand off tasks and share results.
What is the Agent2Agent (A2A) protocol?
A2A, short for Agent2Agent, is an open protocol for communication between AI agents. Google introduced it in April 2025 and later donated it to the Linux Foundation, where a steering committee with companies such as AWS, Cisco, IBM, Microsoft, Salesforce, SAP and ServiceNow maintains it. Its goal is that an agent built with one framework or vendor can work with agents built with any other.
Each agent publishes an Agent Card, a JSON document that describes what it can do, where to reach it and how to authenticate. A client agent sends it a message; the remote agent tracks the work as a task that moves through states such as working, input required, completed or failed, and returns its results as artifacts made of text, files or structured data. A2A runs over standard web technology: JSON-RPC 2.0, gRPC or plain HTTP with JSON, with Server-Sent Events for streaming updates and webhooks for long-running work.
If MCP is how an agent picks up its tools, A2A is how it calls a colleague: it doesn't need to know how the other agent works inside, only what it offers and how to ask.
A2A is often presented as a rival to MCP, but the two solve different problems and are designed to work together: MCP connects an agent to tools and data, while A2A connects independent agents to each other. A travel assistant might use MCP to search flights through an airline's API and A2A to hand the hotel booking to a hotel chain's own agent.
Key takeaways
- A2A is an open protocol that lets AI agents talk to each other.
- Each agent describes itself in an Agent Card: its skills, its address and its authentication.
- Work is tracked as tasks with states, and results come back as artifacts.
- It uses JSON-RPC 2.0, gRPC or HTTP, with streaming over Server-Sent Events.
- MCP connects agents to tools; A2A connects agents to agents.
Readers ask
What is the difference between A2A and MCP?
MCP connects one AI agent to tools, data and prompts. A2A lets separate agents, possibly from different companies, discover each other and delegate work. Many systems use both.
What is an Agent Card?
A JSON document that an A2A agent publishes to describe itself: its name, what it can do, how to reach it and which authentication it requires. Other agents read it to decide whether and how to call it.
Who maintains A2A?
It started at Google and is now an open-source project at the Linux Foundation, run by a technical steering committee with members from several large technology companies.
See also
- MCP vs A2AMCP connects an AI agent to tools, data and prompts, while A2A lets separate AI agents, even from different vendors, find each other and hand off tasks.
- Model Context ProtocolAI & Machine Learning, p. 40The Model Context Protocol is an open standard that defines how AI applications connect to external tools, data sources, and prompts through a shared interface.
- AI AgentAI & Machine Learning, p. 3An AI agent is a system that uses an LLM to plan and carry out multi-step tasks by deciding which tools to call, observing the results, and acting again.
- Tool CallingAI & Machine Learning, p. 61Tool calling is an LLM feature in which the model asks the application to run a specific function with structured arguments, then uses the result in its answer.
- RPCBackend & APIs, p. 43RPC is a communication style in which a program calls a function that runs on another machine as if it were a local function, hiding the network in between.
- APIBackend & APIs, p. 2An API is a set of rules that lets one piece of software request data or actions from another in a predictable, documented way.
- JSONBackend & APIs, p. 28JSON is a lightweight, text-based format for storing and exchanging structured data as key-value pairs and lists, readable by both humans and machines.
Free · once a week
Software and AI, explained once a week
The week's software and AI news with every term explained, plus a term to learn and a quick quiz. A two-minute read.
See this week's issueAbout this page
Spotted a mistake or something missing on this page?Suggest an edit
How pages are made