A2A MCP
Enables Claude to connect and interact with A2A Protocol agents, allowing discovery of agent capabilities, sending messages to remote agents, managing multi-turn conversations, and viewing artifacts returned by agents.
What is A2A MCP?
A2A MCP is a Model Context Protocol (MCP) server that allows AI assistants like Claude, Cursor, and VS Code to enables claude to connect and interact with a2a protocol agents, allowing discovery of agent capabilities, sending messages to remote agents, managing multi-turn conversations, and viewing artifacts r...
Enables Claude to connect and interact with A2A Protocol agents, allowing discovery of agent capabilities, sending messages to remote agents, managing multi-turn conversations, and viewing artifacts returned by agents.
This server falls under the Coding Agents category on MCPgee, the world's largest MCP server directory with 33,000+ servers.
Features
- Enables Claude to connect and interact with A2A Protocol age
Use Cases
Maintainer
Works with
Installation
Manual Installation
npx a2a-mcp-serverConfiguration
Configuration Details
claude_desktop_config.json
Performance
Response Metrics
Resource Usage
How to Set Up and Use A2A MCP
The A2A MCP Server acts as a protocol bridge that lets MCP clients like Claude connect to and communicate with remote agents implementing Google's Agent-to-Agent (A2A) protocol. It enables Claude to register A2A agents by URL, discover their capabilities, send messages both synchronously and as streaming responses, retrieve task results, and cancel running tasks — all without the MCP client needing to understand A2A protocol details. Developers building multi-agent systems can use it to connect Claude to specialized A2A agents for code review, research, data processing, or any other capability exposed through the A2A standard.
Prerequisites
- Python 3.10+ with pip available
- At least one running A2A-compatible agent accessible at a known URL (e.g., http://localhost:41242)
- pip install a2a-mcp-server, or Smithery CLI for automatic install
- An MCP client such as Claude Desktop or Claude Code
Install the server via pip
Install the a2a-mcp-server package from PyPI. This provides the server binary and all required A2A protocol dependencies.
pip install a2a-mcp-serverAlternatively, install via Smithery
If you use Smithery CLI for MCP server management, it can install and configure the server automatically for the Claude client.
npx -y @smithery/cli install @GongRzhe/A2A-MCP-Server --client claudeAdd to your MCP client configuration
Register the server in your MCP client config. The server runs over stdio by default, which works with all MCP clients. Optional environment variables control transport mode if you need HTTP/SSE instead.
{
"mcpServers": {
"a2a-mcp": {
"command": "python",
"args": ["-m", "a2a_mcp_server"]
}
}
}Start your A2A agent(s)
Ensure any A2A-compatible agents you want to connect to are running and accessible at their endpoint URLs before registering them. A2A agents expose an agent card at their root URL describing their capabilities.
Register an agent and start communicating
Through Claude, register your A2A agent by URL, then send it a message to discover its capabilities or assign it a task.
A2A MCP Examples
Client configuration
Example claude_desktop_config.json entry for the A2A MCP server using stdio transport.
{
"mcpServers": {
"a2a-mcp": {
"command": "python",
"args": ["-m", "a2a_mcp_server"],
"env": {
"MCP_TRANSPORT": "stdio"
}
}
}
}Prompts to try
Sample requests using the server's agent management and communication tools.
- "Register the A2A agent at http://localhost:41242 and tell me what it can do."
- "List all registered A2A agents."
- "Send the message 'Analyze this dataset and return a summary' to the agent at http://localhost:41242."
- "Get the result for task ID abc123 from the A2A agent."
- "Cancel the currently running task on the data processing agent."Troubleshooting A2A MCP
register_agent fails with a connection refused error
Verify the A2A agent is running and listening at the URL you provided. Test by fetching the agent card manually: 'curl http://localhost:41242/' should return a JSON description of the agent's capabilities.
get_task_result returns 'task not found' after send_message
Use the task ID returned by send_message — not a manually constructed one. A2A tasks are identified by UUIDs assigned at creation. If the agent is slow, wait a moment before calling get_task_result, or use send_message_stream for real-time output.
Server fails to start with ImportError or ModuleNotFoundError
Ensure the package installed successfully with 'pip show a2a-mcp-server'. If modules are missing, the installation may have been incomplete. Try reinstalling with 'pip install --force-reinstall a2a-mcp-server' in a clean virtual environment.
Frequently Asked Questions about A2A MCP
What is A2A MCP?
A2A MCP is a Model Context Protocol (MCP) server that enables claude to connect and interact with a2a protocol agents, allowing discovery of agent capabilities, sending messages to remote agents, managing multi-turn conversations, and viewing artifacts returned by agents. It connects AI assistants to external tools and data sources through a standardized interface.
How do I install A2A MCP?
Follow the installation instructions on the A2A MCP GitHub repository. Clone the repo, install dependencies, and add the server config to your AI client.
Which AI clients work with A2A MCP?
A2A MCP works with all major MCP-compatible AI clients including Claude Desktop, Claude Code, Cursor, VS Code (GitHub Copilot), Windsurf, and Cline.
Is A2A MCP free to use?
Yes, A2A MCP is open source and available under the Apache 2.0 license. You can use it freely in both personal and commercial projects.
A2A MCP Alternatives — Similar Coding Agents Servers
Looking for alternatives to A2A MCP? Here are other popular coding agents servers you can use with Claude, Cursor, and VS Code.
Dify
★ 142.2kProduction-ready platform for agentic workflow development.
Ruflo
★ 54.0k🌊 The leading agent orchestration platform for Claude. Deploy intelligent multi-agent swarms, coordinate autonomous workflows, and build conversational AI systems. Features enterprise-grade architecture, self-learning swarm intelligence, RAG integrat
Goose
★ 45.7kan open source, extensible AI agent that goes beyond code suggestions - install, execute, edit, and test with any LLM
Antigravity Awesome Skills
★ 38.3kInstallable GitHub library of 1,400+ agentic skills for Claude Code, Cursor, Codex CLI, Gemini CLI, Antigravity, and more. Includes installer CLI, bundles, workflows, and official/community skill collections.
AgentScope
★ 25.5kBuild and run agents you can see, understand and trust.
Serena
★ 24.5kA coding agent toolkit that provides IDE-like semantic code retrieval and editing tools, enabling LLMs to efficiently navigate and modify codebases using symbol-level operations instead of basic file reading and string replacements.
Browse More Coding Agents MCP Servers
Explore all coding agents servers available in the MCPgee directory. Each server includes setup guides for Claude, Cursor, and VS Code.
Set Up A2A MCP in Your Editor
Choose your AI client for step-by-step setup instructions.
Quick Config Preview
Add this to your claude_desktop_config.json or .cursor/mcp.json
Ready to use A2A MCP?
Browse our complete directory of 33,000+ MCP servers, read setup guides for your editor, and start building with the Model Context Protocol.