Overview
Coursera Flash Sale
40% Off Coursera Plus for 3 Months!
Grab it
Learn to build sophisticated AI agents that communicate and collaborate using Agent-to-Agent (A2A) communication Protocol and Model Context Protocol (MCP) in this 25-minute technical tutorial from Google. Discover how these open standards enable AI agents to work together securely and efficiently to solve complex problems. Explore the architecture behind agent communication and follow along with practical code examples using Google's Agent Development Kit (ADK) to create specialized agents that can delegate tasks and collaborate effectively. Walk through defining both client and remote A2A agents, understand A2A Agent Cards, learn how to host remote agents, and see how to expose inventory tools as an MCP server. Watch a live demonstration of the concepts in action and get answers to community questions about implementing these protocols in your own projects.
Syllabus
0:00 - Introduction
1:10 - Agent usecase
2:19 - What is A2A protocol?
3:02 - What is MCP?
4:06 - Architecture
5:14 - Code walkthrough: Define agents
6:18 - Define client A2A agent
9:00 - Define remote A2A agent
11:06 - A2A Agent Card
12:57 - How to host remote agent?
13:25 - How to expose inventory tools as an MCP server
15:51 - Demo
22:20 - Community questions
24:18 - Outro and next steps
Taught by
Google Developers