Agent-to-Agent with A2A
How agents talk to each other beyond tools.
Agent-to-Agent with A2A is a free MCP Academy lesson on CoddyKit — lesson 3 of 4. You can read the complete lesson below for free — then practise it hands-on in the browser with a built-in code editor and a 24/7 AI tutor. It is part of the MCP Academy learning path, one of 4 lessons in the course, and your progress syncs across the web and the CoddyKit app.
Beyond Tools
MCP connects a model to tools and data. But what if one agent needs to talk to another whole agent? That is a different problem.
Meet A2A
A2A stands for Agent-to-Agent. It is an open protocol for agents to discover each other and collaborate as peers, not as tools.
Tool vs Teammate
MCP treats a server as a capability you call. A2A treats the other side as a teammate that can reason and respond on its own.
The Agent Card
An A2A agent publishes an Agent Card, a small description of who it is and what it can do, so others can find and trust it.
Tasks, Not Tool Calls
In A2A you send a task and the other agent works on it, possibly over many turns, then hands back a result or an update.
Long-Running Work
A2A is built for jobs that take time. An agent can report progress and stream partial results instead of one quick reply.
Complementary, Not Rival
MCP and A2A solve different layers. One agent can use MCP for its tools while using A2A to delegate to another agent.
A Quick Picture
Picture a planner agent that calls a weather tool over MCP, then asks a booking agent over A2A to reserve the trip.
Opaque by Design
A2A agents stay opaque: you do not see their inner tools or prompts, only the task you send and the result you get back.
Open and Vendor-Neutral
Like MCP, A2A is an open standard, so agents built by different teams or vendors can still cooperate over it.
Choosing Between Them
Need a function or data source? Use MCP. Need to hand a whole goal to another autonomous agent? Reach for A2A.
Quick Check
Tell A2A and MCP apart by what they connect.
Recap
A2A links agents as peers that exchange tasks; MCP links an agent to tools. Use both together for richer systems. 🤝
Frequently asked questions
Is the “Agent-to-Agent with A2A” lesson free?
Yes — the full text of “Agent-to-Agent with A2A” is free to read here on the web, and the MCP Academy course includes 4 lessons in total. To practise it interactively (a built-in code editor and a 24/7 AI tutor) and unlock the rest of the MCP Academy course, upgrade to CoddyKit PRO.
What will I learn in “Agent-to-Agent with A2A”?
How agents talk to each other beyond tools. You practise MCP Academy with hands-on code you run directly in the browser, and a 24/7 AI tutor answers your questions as you work through the lesson.
Do I need any experience to start MCP Academy?
No prior experience is required. MCP Academy on CoddyKit is structured for beginners through advanced learners; this is — lesson 3 of 4, so you can start here or from the beginning and move at your own pace.
How long does the “Agent-to-Agent with A2A” lesson take?
Most CoddyKit lessons take about 5–10 minutes. Each one is bite-sized and interactive, so you make steady progress and pick up exactly where you left off across the web and the app.
Can I write and run code in this MCP Academy lesson?
Yes. Every MCP Academy lesson includes a built-in code editor, so you write and run real code right in your browser and get instant AI feedback — no local setup required.
All lessons in this course
- MCP vs Function Calling
- MCP Inside Agent Frameworks
- Agent-to-Agent with A2A
- Where MCP Is Heading Next