A2A Protocol
~3 min read
Concept & How It Works
Why Does It Exist?
Siloed agents can't collaborate across vendors. A2A provides a common wire format so your LangGraph agent can delegate to a partner's ADK agent without custom integration per pair.
Real-World Analogy
A2A is HTTP for agents — a universal request/response contract so any agent can call any other agent that speaks the protocol.
Visual Workflows
What is A2A Protocol?
Example
Scenario
Your planner agent discovers a `research-agent` card, sends a Task to summarize a PDF, receives streaming status updates, and gets the final Report artifact.
Solution
In Multi-Agent Systems, apply A2A Protocol to this scenario: Your planner agent discovers a `research-agent` card, sends a Task to summarize a PDF, receives streaming status updates, and gets the final Report artifact. Identify the inputs, run the technique, validate the output, and note one thing you would monitor in production.
Practice Task
Do this before moving to the next module — reading alone is not enough.
Open the Code Walkthrough below and run it locally. Change one parameter related to A2A Protocol (e.g. model, temperature, top_k, or tool name), observe the difference in output, and write 2–3 sentences explaining what changed.
Code Walkthrough
Highlighted lines show where A2A Protocol happens in the code.
1// Agent Card snippet2{3 "name": "research-agent",4 "skills": [{"id": "summarize", "description": "Summarize documents"}],5 "url": "https://research.example.com/a2a"6}7// Task send: POST with method tasks/send, params: { message, skill }Commands to Remember
Commands to Remember
pip install langgraph langchain-openai # multi-agent orchestrationpip install crewai # role-based multi-agent crews
Common Mistakes
- Treating A2A Protocol as a black box without evaluation
- Ignoring cost and latency in production
- Skipping error handling for a2a protocol
Cheat Sheet
Quick recap — the most important points from this module.
Cheat Sheet
quick ref- •A2A Protocol
- •Agent Card
- •Task
- •JSON-RPC