What is A2a Agent Mcpserver Generator

A2A Agent to MCP server is a command-line tool designed to generate Model Context Protocol (MCP) server code from an A2A agent. It serves as a bridge between Large Language Models (LLMs) and the A2A agent, simplifying the process of creating an MCP server.

Use cases

This tool can be used to quickly set up an MCP server for applications that require interaction between A2A agents and LLMs, making it easier to implement AI-driven solutions in various domains such as chatbots, virtual assistants, and automated customer support systems.

How to use

To use the tool, first start an A2A agent instance. Then execute the command ‘uv run a2a-agent-mcpserver-generator --agent <agent_url>’ to generate the necessary MCP server code based on the skills of your A2A agent. You can configure various options including output directory, server name, and more.

Key features

Key features include automatic generation of MCP tools from A2A agent skills, support for standard input/output transport, complete project setup with all necessary files, and easy configuration through environment variables, allowing for flexibility in running the MCP server.

Where to use

The generated MCP server can be used in environments that require seamless communication between A2A agents and LLMs. Ideal applications include integrating AI functionalities into existing software systems, enhancing conversational agents, or developing new AI applications that leverage A2A agents for improved responsiveness and capabilities.

A2A Agent to MCP server 🤖

A command-line tool that generates Model Context Protocol (MCP) server code from a oneline a2a agent. This tool helps you quickly create an MCP server that acts as a bridge between LLMs (Large Language Models) and the a2a agent.

English | 简体中文

Features ✨

• Automatic Tool Generation: Converts each skills of the a2a agent into an MCP tool
• Transport Options: Only supports stdio, for sse you can leveral mcp-proxy
• Complete Project Setup: Generates all necessary files to run an MCP server
• Easy Configuration: Simple environment-based configuration for the generated server

Prerequisites

• Python 3.13+
• uv (optional, but recommended) or pip

Environment Setup 🔧

Firstly, you need to start up a a2a agent:
https://github.com/google/a2a-python/tree/main/examples

uv run a2a-agent-mcpserver-generator --agent http://0.0.0.0:10000

Installation 📦

uv pip install a2a-agent-mcpserver-generator

Usage 🚀

uv run a2a-agent-mcpserver-generator --help
Usage: a2a-agent-mcpserver-generator [OPTIONS]

Options:
  --agent TEXT
  --output TEXT
  --name TEXT
  --history BOOLEAN
  --use_push_notifications BOOLEAN
  --push_notification_receiver TEXT
  --help                          Show this message and exit.

E2E example

Suggest use mcpclihost as MCP host to take a try.
This tool(mcpclihost) could support both Azure Openai and deepseek

You can add generated MCP server congiguration like this:

{
  "mcpServers": {
    "a2a-mcp": {
      "command": "uv",
      "args": [
        "--project",
        "/Users/lipeng/workspaces/github.com/vincent-pli/a2a-agent-mcpserver-generator/a2a-mcp-server",
        "run",
        "a2a-agent-mcpserver"
      ]
    }
  }
}

to the ~/.mcp.json(default mcp server configuration path of mcpclihost), then take a try

License 📄

This project is licensed under the Apache 2.0 License - see the LICENSE file for details.