Skip to main content
Glama
AlwaysSany

Multi-Server MCP FastAPI Project

by AlwaysSany
OverviewSchemaRelated ServersScoreDiscussions
Python
Remote

Multi-Server MCP FastAPI Project

A comprehensive FastAPI application that hosts multiple Model Context Protocol (MCP) servers, each providing access to different public APIs and services.

Work Demo

fastmcp-multi-server-mcp.webm

🚀 Features

This project includes 5 different MCP servers:

  • Weather Server – OpenWeatherMap API integration for weather data

  • News Server – NewsAPI integration for latest news articles

  • Currency Server – Exchange rates and currency conversion

  • Quote Server – Inspirational quotes and random facts

📋 Prerequisites

  • Python 3.13+

  • UV package manager

  • API keys for external services (see Configuration section)

📁 Project Structure

fastapi-multi-server-mcp/
├── main.py                 # FastAPI application entry point
├── requirements.txt        # Python dependencies
├── pyproject.toml         # UV project configuration
├── .env.example           # Environment variables template
├── README.md              # This file
├── servers/               # MCP servers directory
│   ├── __init__.py
│   ├── weather_server.py  # Weather API MCP server
│   ├── news_server.py     # News API MCP server
│   ├── currency_server.py # Currency API MCP server
│   └── quote_server.py    # Quotes API MCP server
├── utils/                 # Utility modules
│   ├── __init__.py
│   ├── api_clients.py     # HTTP client utilities
│   └── config.py          # Configuration management
├── tests/                 # Test suite
│   ├── __init__.py
│   ├── test_weather.py
│   ├── test_news.py
│   ├── test_currency.py
│   └── test_quote.py
└── docs/                  # Documentation
    ├── api_reference.md
    └── deployment.md

🛠️ Installation

Using UV (Recommended)

# Clone the repository
git clone <your-repo-url>
cd fastapi-multi-server-mcp

# Create virtual environment and install dependencies
uv venv
source .venv/bin/activate  # On Windows: .venv\Scripts\activate
uv pip install -r requirements.txt

Using pip

pip install -r requirements.txt

🔧 Configuration

Create a .env file in the root directory with your API keys:

# Weather API (OpenWeatherMap)
OPENWEATHER_API_KEY=your_openweather_api_key

# News API
NEWS_API_KEY=your_news_api_key


# Exchange Rates API
EXCHANGE_RATES_API_KEY=your_exchange_rates_api_key

# Server Configuration
PORT=10000
HOST=0.0.0.0

Getting API Keys

🏃‍♂️ Running the Application

Development

# Using UV
uv run uvicorn main:app --reload --host 0.0.0.0 --port 10000 --reload

# Using `npx` to debug the running application,
npx @modelcontextprotocol/inspector uv run uvicorn main:app

To connect specific server to inspector,

Set,

  1. Transport Type: StreamableHTTP

  2. URLs:

  • for news: http://localhost:10000/news/mcp/

  • for weather: http://localhost:10000/weather/mcp/

  • for currency: http://localhost:10000/currency/mcp/

  • for quotes: http://localhost:10000/quotes/mcp/

  1. Click connect, it will open inspector window like this:

Quotes server:

Inspector

News server:

Inspector

Weather server:

Inspector

Currency server:

Inspector

As of June 2025, Windsurf and Cursor supports two transport types for MCP servers: stdio and /sse, but not StreamableHTTP. So using stdio or /sse as transport type is recommended. But with inspector, you can connect to the MCP server running on FastAPI with any transport type.

Production

# Using Gunicorn
gunicorn main:app -w 4 -k uvicorn.workers.UvicornWorker --bind 0.0.0.0:10000

📚 API Endpoints

Once running everything successfully,

Your FastAPI server will be available at: http://localhost:10000, docs at http://localhost:10000/docs and your MCP servers will be available at:

  • Weather Server: http://localhost:10000/weather

  • News Server: http://localhost:10000/news

  • Currency Server: http://localhost:10000/currency

  • Quote Server: http://localhost:10000/quotes

🧪 Testing

# Run all tests
pytest

# Run with coverage
pytest --cov=.

# Run specific server tests
pytest tests/test_weather.py

🛠️ Available Tools by Server

Weather Server (/weather)

get_current_weather - Get current weather for a city
get_weather_forecast - Get 5-day weather forecast
get_weather_by_coordinates - Get weather by latitude/longitude

News Server (/news)

get_top_headlines - Get top news headlines
search_news - Search news articles by keyword
get_news_by_category - Get news by category (business, tech, sports, etc.)

Currency Server (/currency)

get_exchange_rates - Get current exchange rates
convert_currency - Convert between currencies
get_supported_currencies - List all supported currencies

Quote Server (/quotes)

get_random_quote - Get a random inspirational quote
get_quote_by_category - Get quotes by category
get_random_fact - Get a random interesting fact

🚀 Deployment

Docker

You can deploy this project using Docker. Example Dockerfile is included:

FROM python:3.11-slim
WORKDIR /app
COPY requirements.txt .
RUN pip install -r requirements.txt
COPY . .
EXPOSE 10000
CMD ["uvicorn", "main:app", "--host", "0.0.0.0", "--port", "10000"]

Environment Variables for Production

PORT=10000
HOST=0.0.0.0
LOG_LEVEL=info
WORKERS=4

📝 License

This project is licensed under the MIT License - see the LICENSE.md file for details.

📚 Support & Documentation

  • Check the documentation in the docs/ directory

  • Review the test files for usage examples

🔄 Updates and Roadmap

  • Add WebSocket support for real-time data

  • Implement caching for API responses

  • Add rate limiting and authentication

  • Create web dashboard for monitoring

  • Add more API integrations (Spotify, Twitter, etc.)

This server cannot be installed

A
license - permissive license
-
quality - not tested
C
maintenance

How are these scores calculated?

Resources

Unclaimed servers have limited discoverability.

Looking for Admin?

If you are the server author, to access and configure the admin panel.

Latest Blog Posts

MCP directory API

We provide all the information about MCP servers via our MCP API.

curl -X GET 'https://glama.ai/api/mcp/v1/servers/AlwaysSany/fastapi-multi-server-mcp'

If you have feedback or need assistance with the MCP directory API, please join our Discord server