FastMCP-ADK-Server-Client

This Repo contain FASTMCP client and FASTMCP server code. FASTMCP Server contain ADK agent

---

# FastMCP-ADK-Server-Client

A demo project that combines **FastMCP** and **Google’s Agent Development Kit (ADK)** to build a powerful MCP (Model Context Protocol) server and client setup.

---

## Overview

This repository includes:

- **server.py** — A FastMCP-based server, wrapping ADK agents to expose tools via MCP.
- **client.py** — A FastMCP client to interact with the MCP server programmatically.
- **requirements.txt** — Python dependencies needed for setup.
- An optional `.env` for configuration (e.g. keys, endpoints).

---

## Prerequisites

- Python 3.10+  
- `uv` package manager (optional, but recommended for easier dependency handling)

---

## Installation

```bash
# Clone the repository
git clone https://github.com/shdhumale/FastMCP-ADK-Server-Client.git
cd FastMCP-ADK-Server-Client

# Using uv (recommended)
uv run install

# Or with pip directly
pip install -r requirements.txt

---

Usage

1. Starting the MCP Server

python server.py

This launches your FastMCP server bundled with ADK agents. The server will begin listening for incoming MCP requests.

2. Running the Client

python client.py

This will initiate the FastMCP client, which can:

• Connect to the MCP server
• List available tools, resources, or prompts
• Send requests to the server (e.g., invoking a tool)

---

Project Structure

├── client.py            # FastMCP client implementation
├── server.py            # FastMCP server implementing ADK agents
├── requirements.txt     # Python dependencies
└── .env                 # Optional config file for API keys or endpoints

---

How It Works

• FastMCP simplifies building MCP servers & clients with a Pythonic decorator-based interface (GitHub, FastMCP).
• ADK (Agent Development Kit) lets you embed intelligent agent logic—tools, session state, event handling—into the MCP server.

When combined, this repo enables:

1. Declarative creation of tools via FastMCP in server.py.
2. High-level client interaction in client.py.
3. Modular, agentic behavior through ADK integration.

---

Notes & Tips

• Use virtual environments to isolate dependencies.
• Populate .env for any required API credentials or environment configuration.
• Depending on your setup, you might add documentation or logging for better traceability.

---

Contributing

You're welcome to contribute enhancements! Here are a few ideas:

• Add new ADK agents or MCP tools.
• Improve command-line flags or configuration options.
• Include better logging, error-handling, or testing.

To contribute:

1. Fork the repo.
2. Create your branch (e.g., feature/new-tool).
3. Commit and push your changes.
4. Submit a Pull Request.

---

License

• (Your chosen license—e.g., MIT, Apache-2.0, etc.) *

---

Acknowledgements

• Built with FastMCP — the high-level MCP framework for Python (GitHub).
• Leveraging Google’s ADK for intelligent agent integration.

---