MCP-Servers

A collection of Model Context Protocol (MCP) servers for AI assistant integration.

Overview

This repository contains multiple MCP server implementations that extend AI assistants with various capabilities:

Project	Description	Language
fastmcp	Comprehensive framework for building MCP servers and clients	Python 3.10+
mcp_youtube_dlp	YouTube video and audio download tools	Python 3.13+
weather	Weather information retrieval	Python

What is MCP?

The Model Context Protocol (MCP) is an open standard created by Anthropic that provides a universal way to connect AI assistants to external data sources and tools. It uses JSON-RPC 2.0 for communication and supports multiple transport protocols (stdio, HTTP).

For detailed technical information, see MCP-web-research.md.

Quick Start

Prerequisites

• Python 3.10+ (3.13+ for mcp_youtube_dlp)
• uv package manager
• Git

Installation

1. Clone this repository:

   git clone https://github.com/yourusername/MCP-Servers.git
   cd MCP-Servers

2. Install dependencies for a specific project:

   cd fastmcp  # or mcp_youtube_dlp, weather
   uv sync

Projects

FastMCP

A comprehensive Python framework for building MCP servers and clients. Includes tools, resources, prompts, authentication, and multiple transport options.

cd fastmcp
uv sync
uv run fastmcp run examples/echo.py

See fastmcp/AGENTS.md for development guidelines.

MCP YouTube-DLP

Download YouTube videos and audio using yt-dlp through MCP tools.

cd mcp_youtube_dlp
uv pip install -e .
python main.py

Environment Variables:

• YT_DLP_PATH - Path to yt-dlp executable
• DEFAULT_DOWNLOAD_DIR - Download directory

Weather

Weather information MCP server.

cd weather
uv sync
python src/weather/server.py

MCP Client Configuration

Add servers to your MCP client (e.g., Claude Desktop):

{
  "mcpServers": {
    "youtube": {
      "command": "python",
      "args": ["/path/to/mcp_youtube_dlp/main.py"],
      "env": {
        "YT_DLP_PATH": "/usr/local/bin/yt-dlp"
      }
    },
    "weather": {
      "command": "python",
      "args": ["/path/to/weather/src/weather/server.py"]
    }
  }
}

Development

All projects use uv as the package manager. Do not use pip directly.

# Install dependencies
uv sync

# Run tests (FastMCP)
uv run pytest

# Lint and format (FastMCP)
uv run pre-commit run --all-files

See CLAUDE.md for detailed development guidelines.

Documentation

• CLAUDE.md - Development guidelines for Claude Code
• MCP-web-research.md - Comprehensive MCP technical reference

License

• fastmcp - MIT License
• mcp_youtube_dlp - GPL-2.0 License
• weather - See project directory

Resources

• MCP Official Documentation
• MCP Python SDK
• FastMCP Documentation