Spotify Playlist Curator MCP Server

Spotify Playlist Curator MCP Server

An MCP server that helps curate Spotify playlists by analyzing your existing tracks and using Claude AI to recommend songs based on mood, vibe, BPM, and other musical attributes.

Features

• Connect to your Spotify account and access your playlists
• Analyze the audio features of tracks in your playlists
• Generate a summary of playlist mood, energy, tempo, and other characteristics
• Get song recommendations from Claude AI based on the playlist analysis
• Search for tracks on Spotify
• Add recommended tracks to your playlists
• Create new playlists

Setup

Prerequisites

• Node.js (v14 or higher)
• A Spotify Developer account and registered application
• Claude access via MCP (Machine Conversation Protocol)

Installation

1. Clone this repository:

   git clone https://github.com/lechiffre1/Spotify-Playlist-Curator-MCP-Server.git
   cd Spotify-Playlist-Curator-MCP-Server
   

2. Install dependencies:

   npm install
   

3. Create a .env file based on the provided .env.example:

   cp .env.example .env
   

4. Set up your Spotify Developer credentials:

   • Go to Spotify Developer Dashboard
   • Create a new application
   • Add http://localhost:3000/callback as a Redirect URI
   • Copy your Client ID and Client Secret to the .env file

5. Start the server:

   npm start
   

Usage

Authentication

When you start the server, you'll need to authenticate with Spotify first:

1. Visit http://localhost:3000/login in your browser
2. Log in with your Spotify account and authorize the application
3. After successful authentication, you can close the browser window and return to your MCP client

MCP Methods

The following MCP methods are available:

getPlaylists

Returns a list of your Spotify playlists.

const response = await client.getPlaylists();

getPlaylistDetails

Gets detailed information about a specific playlist, including track analysis.

const response = await client.getPlaylistDetails({
  playlistId: "your_playlist_id"
});

getClaudeRecommendations

Gets song recommendations from Claude AI based on playlist analysis.

const response = await client.getClaudeRecommendations({
  playlistId: "your_playlist_id",
  count: 10 // Optional, defaults to 10
});

addRecommendationsToPlaylist

Adds recommended tracks to a playlist.

const response = await client.addRecommendationsToPlaylist({
  playlistId: "your_playlist_id",
  trackUris: ["spotify:track:id1", "spotify:track:id2", ...]
});

searchTracks

Searches for tracks on Spotify.

const response = await client.searchTracks({
  query: "search query",
  limit: 10 // Optional, defaults to 10
});

createPlaylist

Creates a new Spotify playlist.

const response = await client.createPlaylist({
  name: "My New Playlist",
  description: "Created by Spotify Playlist Curator", // Optional
  isPublic: false // Optional, defaults to false
});

Example Client Usage

import { createClient } from '@anthropic-ai/mcp-client';

async function main() {
  // Create MCP client
  const client = createClient({
    serverUrl: 'http://localhost:3000',
    anthropicApiKey: 'your_anthropic_api_key' // Required for Claude integration
  });

  // 1. Get user playlists
  const playlists = await client.getPlaylists();
  console.log('Your playlists:', playlists);

  // 2. Select a playlist and get its details
  const playlistId = playlists.playlists[0].id;
  const playlistDetails = await client.getPlaylistDetails({ playlistId });
  console.log('Playlist summary:', playlistDetails.summary);

  // 3. Get recommendations from Claude
  const recommendations = await client.getClaudeRecommendations({ playlistId });
  console.log('Claude recommendations:', recommendations.claudeRecommendations);

  // 4. Add recommendations to the playlist
  const trackUris = recommendations.claudeRecommendations
    .filter(track => track.matched)
    .map(track => track.uri);
  
  if (trackUris.length > 0) {
    const result = await client.addRecommendationsToPlaylist({ playlistId, trackUris });
    console.log('Added recommendations:', result);
  }
}

main().catch(console.error);

License

MIT