MCP-SERVER

alee2602/MCP-SERVER

3.1

If you are the rightful owner of MCP-SERVER and would like to certify it and/or have it hosted online, please leave a comment on the right or send an email to dayong@mcphub.com.

The MCP Playlist Server is a sophisticated tool for creating and managing music playlists using advanced audio feature analysis and Spotify track data.

Tools
5
Resources
0
Prompts
0

🎵 MCP PLAYLIST SERVER

A Model Context Protocol (MCP) server that provides intelligent playlist curation tools using Spotify track data and audio feature analysis. This server enables AI assistants to create mood-based playlists, find similar songs, analyze audio characteristics, and curate personalized music collections.

🚀 Features

Core MCP Tools

  • create_mood_playlist: Generate playlists based on emotional states (happy, sad, energetic, calm, party, chill)
  • find_similar_songs: Discover songs with similar audio characteristics using cosine similarity analysis
  • analyze_song: Get comprehensive audio feature breakdown for any track
  • create_genre_playlist: Build genre-focused playlists with customizable diversity levels
  • get_dataset_stats: View detailed dataset statistics and insights

Advanced Audio Analysis

The server analyzes multiple sophisticated audio characteristics:

  • Energy: Track intensity and power measurement
  • Valence: Musical positivity spectrum (happiness to sadness)
  • Danceability: Rhythmic suitability for dancing
  • Acousticness: Acoustic vs electronic instrumentation balance
  • Tempo: Beats per minute analysis
  • Speechiness: Spoken word content detection
  • Instrumentalness: Vocal vs instrumental content ratio
  • Liveness: Live performance detection
  • Popularity: Track mainstream appeal metrics

🏗️ Architecture

├── server/
│   ├── main.py          # FastMCP server implementation
│   └── engine.py        # Playlist curation engine with ML algorithms
├── spotify_songs.csv  # Spotify dataset (32K+ songs)
├── requirements.txt     # Python dependencies
└── README.md           # This file

📋 Requirements

  • Python: 3.10 or higher
  • Dataset: Spotify tracks CSV with audio features
  • Dependencies: Listed in requirements.txt

Core Dependencies

  • fastmcp>=1.2.0 - Modern MCP server framework
  • pandas>=2.0.0 - Data manipulation and analysis
  • numpy>=1.24.0 - Numerical computing
  • scikit-learn>=1.3.0 - Machine learning algorithms

🛠️ Installation

1. Clone Repository

git clone https://github.com/alee2602/MCP-SERVER

2. Environment Setup

# Using Anaconda (recommended)
conda create -n mcp-playlist python=3.11
conda activate mcp-playlist

# Or using venv
python -m venv .venv
source .venv/bin/activate  # Linux/Mac
# .venv\Scripts\activate   # Windows

3. Install dependencies

pip install -r requirements.txt

🧪 Run the server

python server/main.py

🔧 Usage with MCP Hosts

Claude Desktop Integration

  1. Add to your claude_desktop_config.json:
{
  "mcpServers": {
    "mcp-playlist": {
      "command": "python",
      "args": ["server/main.py"],
      "cwd": "/absolute/path/to/your/project"
    }
  }
}

  1. Restart Claude Desktop


  2. Below are examples the assistant understands.

Mood Playlist:

  • Create a happy playlist of 30 minutes in the rap genre
  • Make a chill playlist with 10 songs (min popularity 60).

Similar Songs:

  • Give me 7 songs similar to ‘Pillowtalk’ by ZAYN.
  • Find songs like ‘Worldwide’ by Big Time Rush.

Song analysis:

  • Analyze the audio features of ‘Bohemian Rhapsody’ by Queen.

Genre playlist:

  • Create a rock, pop playlist with 12 songs, diversity high

Dataset stats:

  • Show me dataset stats
  • What are the top genres in this dataset

Other MCP Clients

Configure with:

  • Protocol: STDIO
  • Command: python server/main.py
  • Working Directory: Project root

📊 API Examples

  1. Create a mood-based playlist
{
  "method": "tools/call",
  "params": {
    "name": "create_mood_playlist",
    "arguments": {
      "mood": "energetic",
      "size": 15,
      "genre": "rock",
      "min_popularity": 50,
      "duration_minutes": 30
    }
  }
}
  1. Create Genre Playlist
{
  "method": "tools/call",
  "params": {
    "name": "create_genre_playlist",
    "arguments": {
      "genres": ["pop", "edm"],
      "size": 20,
      "diversity": "high"
    }
  }
}
  1. Find similar songs
{
  "method": "tools/call",
  "params": {
    "name": "find_similar_songs",
    "arguments": {
      "song_name": "Blinding Lights",
      "artist": "The Weeknd",
      "count": 8
    }
  }
}
  1. Comprehensive Song Analysis
{
  "method": "tools/call",
  "params": {
    "name": "analyze_song",
    "arguments": {
      "song_name": "Hotel California",
      "artist": "Eagles"
    }
  }
}
  1. Get Dataset Statistics
{
  "method": "tools/call",
  "params": {
    "name": "get_dataset_stats",
    "arguments": {}
  }
}

🔍 Troubleshooting

1. "Dataset empty" Error

  • Verify spotify_songs.csv exists in project root
  • Check file permissions and format
  • Ensure required columns are present

2. "Import Error" Messages

pip install --upgrade fastmcp pandas scikit-learn

Debug mode

# Enable verbose logging
python server/main.py --debug

🙏 Acknowledgments