A2a Directory
What is A2a Directory
A2A Directory (Agent2Agent) is an open protocol developed by Google that enables AI agents to communicate securely and collaborate, breaking down silos between isolated agent systems for complex cross-application automation.
Use cases
Use cases for A2A Directory include automating workflows across different applications, enabling AI agents to collaborate on tasks, and integrating disparate systems for enhanced functionality.
How to use
To use A2A Directory, start by reading the technical documentation and watching demo videos. Clone the official repository, run sample codes, and build your own A2A agents or clients using the provided libraries.
Key features
Key features of A2A Directory include simplicity with HTTP and JSON-RPC, enterprise readiness focusing on security and privacy, asynchronous handling of long-running tasks, modality agnosticism supporting various data types, and opaque execution allowing agents to interact without sharing internal logic.
Where to use
A2A Directory can be used in various fields such as enterprise automation, AI development, and any application requiring secure communication and collaboration between different AI agents.
Clients Supporting MCP
The following are the main client software that supports the Model Context Protocol. Click the link to visit the official website for more information.
Overview
What is A2a Directory
A2A Directory (Agent2Agent) is an open protocol developed by Google that enables AI agents to communicate securely and collaborate, breaking down silos between isolated agent systems for complex cross-application automation.
Use cases
Use cases for A2A Directory include automating workflows across different applications, enabling AI agents to collaborate on tasks, and integrating disparate systems for enhanced functionality.
How to use
To use A2A Directory, start by reading the technical documentation and watching demo videos. Clone the official repository, run sample codes, and build your own A2A agents or clients using the provided libraries.
Key features
Key features of A2A Directory include simplicity with HTTP and JSON-RPC, enterprise readiness focusing on security and privacy, asynchronous handling of long-running tasks, modality agnosticism supporting various data types, and opaque execution allowing agents to interact without sharing internal logic.
Where to use
A2A Directory can be used in various fields such as enterprise automation, AI development, and any application requiring secure communication and collaboration between different AI agents.
Clients Supporting MCP
The following are the main client software that supports the Model Context Protocol. Click the link to visit the official website for more information.
Content
✨ A2A Directory ✨
🌐 English | 中文 | 日本語 | Deutsch
Contents
- 📋 Overview
- 🚀 Getting Started
- 📚 Resources
- 📦 Official Samples
- 🛠️ Tools
- 🤝 Community Implementations
- 🎯 Community Samples
- 👥 Community
- 🤝 Contributing
Overview
A2A (Agent2Agent) is an open protocol from Google enabling AI agents to communicate securely and collaborate. It breaks down silos between isolated agent systems, allowing for complex cross-application automation.
Key Features:
- 🎯 Simple: Uses HTTP, JSON-RPC, SSE
- 🏢 Enterprise Ready: Focuses on security and privacy
- ⚡ Async First: Handles long-running tasks
- 🔄 Modality Agnostic: Supports text, files, forms, streams
- 🔒 Opaque Execution: Agents interact without sharing internal logic
Official Links:
- 🌐 Website: google.github.io/A2A
- 📦 GitHub: https://github.com/google-a2a/A2A
Getting Started
-
Learn the Basics
- 📖 Read the technical documentation
- 🎥 Watch the demo video
-
Run Samples
- 📥 Clone the official repo
- 📝 Follow instructions in
/samples
-
Build Your Own
- 🛠️ Use official libraries or adapt samples
- 🏗️ Start with a basic A2A agent or client
Resources
Official Documentation
Official Samples
Python Sample Collection
Agent Examples (Intelligent Agents Based on Different Frameworks)
| Project Name | Description | Link |
|---|---|---|
| Google ADK | Expense report filling agent, showcasing multi-turn interactions and web form handling | google_adk |
| AG2 + MCP | MCP-enabled agent based on AG2 framework | ag2 |
| Azure AI Foundry | Agent based on Azure AI Foundry services | azureaifoundry_sdk |
| LangGraph | Currency conversion agent with tool usage and streaming updates | langgraph |
| CrewAI | Image generation agent with multi-turn interactions and image transmission | crewai |
| LlamaIndex | File parsing and chat agent with file upload and streaming updates | llama_index_file_chat |
| Marvin | Contact information extraction agent | marvin |
| MindsDB | Enterprise data agent supporting database queries | mindsdb |
| Semantic Kernel | Travel agent based on Microsoft Semantic Kernel framework | semantickernel |
| AutoGen | AutoGen framework example | autogen |
| Hello World | Basic example agent | helloworld |
| Veo Video Generation | Video generation agent | veo_video_gen |
| Analytics | Analytics agent | analytics |
| A2A Telemetry | Telemetry data agent | a2a_telemetry |
| A2A MCP | MCP protocol agent | a2a_mcp |
| Birthday Planner ADK | Birthday planning agent | birthday_planner_adk |
| Headless Agent Auth | Headless agent authentication example | headless_agent_auth |
Host Applications (Client Applications)
| Project Name | Description | Link |
|---|---|---|
| CLI | Command line client | cli |
| Multiagent | Multi-agent orchestrator | multiagent |
Go Language Examples
| Project Name | Description | Link |
|---|---|---|
| Server | Go language A2A server implementation | server |
| Client | Go language A2A client implementation | client |
| Models | Shared data structures | models |
JavaScript/TypeScript Examples
| Project Name | Description | Link |
|---|---|---|
| Movie Agent | Movie information search agent using TMDB API | movie-agent |
| Coder Agent | Code generation agent | coder |
Tools
Essential tools for A2A protocol development, testing, and validation.
Community Implementations
| Name | Author | Description | Stars |
|---|---|---|---|
| a2a-python | Official Python SDK for the Agent2Agent (A2A) Protocol |  |
|
| a2a-js | @google-a2a | Official JavaScript SDK for the Agent2Agent (A2A) Protocol - A JavaScript library that helps run agentic applications as A2AServers |  |
| a2ajava | @vishalmysore | A pure Java implementation of Google’s A2A protocol for Spring Boot applications, featuring both client and server implementations |  |
| a2a4j | @a2ap | A2A4J is a comprehensive Java implementation of the Agent2Agent Protocol, including server, client, examples, and a starter — ready to use out of the box. |  |
| legion-a2a | @TheRaLabs | A TypeScript implementation of the A2A protocol with a focus on modularity and extensibility |  |
| trpc-a2a-go | @trpc-group | Go A2A implementation by the tRPC team featuring full client/server support, in-memory task management, streaming responses, session management, multiple auth methods (JWT, API Key, OAuth2), and comprehensive examples |  |
| jira-a2a | @tuannvm | The Jira A2A system is a DevOps workflow automation platform using the tRPC-A2A-Go framework. It consists of independent Go agents that communicate via A2A messages. |  |
| a2a-go | @a2aserver | A Go library for building A2A servers, with example implementations |  |
| a2a-rs | @EmilLindfors | An idiomatic Rust implementation following hexagonal architecture principles |  |
| a2a_min | @pcingola | A minimalistic Python SDK for A2A communication |  |
| a2adotnet | @azixaka | A C#/.NET implementation of the A2A protocol |  |
| nestjs-a2a | @thestupd | A module for integrating the A2A protocol into NestJS applications |  |
| python-a2a | @themanojdesai | An easy-to-use Python library for implementing the A2A protocol |  |
| Aira | @IhateCreatingUserNames2 | An A2A network implementation for hosting, registering, discovering, and interacting with agents |  |
| Cognisphere | @IhateCreatingUserNames2 | An AI agent development framework built on Google’s ADK, facilitating agent creation potentially for A2A networks |  |
| a2a-server | @chrishayuk | A lightweight A2A python implementation |  |
| a2a-cli | @chrishayuk | A command-line client for the A2A |  |
| A2A Test Suit | @robert-at-pretension-io | A2A Test Suite |  |
| Grasp | @adcentury | A Self-hosted Browser Using Agent with built-in MCP and A2A support |  |
| swissknife | @daltonnyx | A multi-agent chat application with MCP support, aiming to expose agents via the A2A protocol and connect to remote A2A agents as a client |  |
| artinet-sdk | @the-artinet-project | A JS/TS SDK for the Agent2Agent Protocol with a focus on developer experience and comprehensive features |  |
| a2a-validation-tool | @llmx-de | A desktop application for testing and validating Agent-to-Agent (A2A) protocol implementations |  |
Community Samples
| Name | Author | Description | Stars |
|---|---|---|---|
| a2a-agent-coder | @sing1ee | A Coder Agent implementation with A2A Server and Client |  |
| agentic-trading | @kweinmeister | A sample application demonstrating Google ADK and A2A interoperability for trading automation |  |
| python-a2a-tutorial | @sing1ee | A comprehensive tutorial for implementing A2A in Python with practical examples |  |
| a2a-python-currency | @sing1ee | A tutorial implementation of a Currency Agent using the A2A Python SDK |  |
| a2a-mcp-openrouter | @sing1ee | Demonstrates A2A + MCP integration using OpenRouter as LLM provider, showcasing unified interface for agent-to-agent communication and tool invocation |  |
| a2a_llama_index_file_chat | @sing1ee | A LlamaIndex-based file chat agent supporting file upload/parsing, conversational interactions, streaming responses, and in-line citations |  |
Community
Contributing
Contributions are welcome! Please read the contributing guidelines first.
Dev Tools Supporting MCP
The following are the main code editors that support the Model Context Protocol. Click the link to visit the official website for more information.
