Enables Siri voice command integration through Apple Shortcuts via an HTTP bridge, allowing voice-controlled Spotify operations on iOS devices.
Provides comprehensive Spotify control including playlist/album/track playback, music search, playback control (play, pause, skip, volume), current track information, and sleep timer functionality.
Spotify MCP Server for Gemini
A Model Context Protocol (MCP) server that exposes Spotify functionality for seamless integration with Google Gemini models.
Features
Play Control: Play playlists, albums, and tracks
Search: Find music by song, artist, or album
Playback Control: Play, pause, skip, and adjust volume
Sleep Timer: Automatically pause playback after a specified duration
Siri Integration: HTTP bridge for Apple Shortcuts (optional)
Prerequisites
Node.js 18+
npm or yarn
Spotify Developer account
TypeScript 5.3+
Google Gemini API access (or compatible Gemini client)
Setup
1. Spotify Developer Setup
Create a new app
Note your Client ID and Client Secret
Add a redirect URI:
http://localhost:3000/callback(or your preferred URI)
2. Project Setup
Clone or download this repository
Install dependencies:
npm installCreate a
.envfile in the root directory:SPOTIFY_CLIENT_ID=your_client_id_here SPOTIFY_CLIENT_SECRET=your_client_secret_here SPOTIFY_REDIRECT_URI=http://localhost:3000/callback HTTP_BRIDGE_PORT=3001 HTTP_BRIDGE_API_KEY=your_optional_api_key_hereBuild the project:
npm run build
3. Authentication
Start the server:
npm startThe server will guide you through the OAuth flow if you're not authenticated
Visit the provided authorization URL
Authorize the app and copy the callback code
The tokens will be saved automatically in
tokens.json
Usage
MCP Server for Gemini
The MCP server runs on stdio and communicates with Gemini via the Model Context Protocol.
Quick Setup:
Configure your Gemini client to use the MCP server (see MCP_SETUP.md for details)
Point it to:
node /path/to/SpotifyMCP/dist/server.jsSet environment variables for Spotify credentials
For detailed setup instructions, see
Run the server directly:
HTTP Bridge (for Siri/Shortcuts)
The HTTP bridge exposes REST endpoints for Apple Shortcuts:
MCP Tools
The server exposes the following MCP tools for Gemini:
play_playlist- Play a playlist by nameplay_album- Play an album by nameplay_track- Play a track by namesearch_music- Search for musiccontrol_playback- Control playback (play, pause, skip, volume)get_current_playing- Get currently playing trackset_sleep_timer- Set a sleep timercancel_sleep_timer- Cancel active timersget_active_timers- List active timers
Example Gemini Interactions
Once configured, you can ask Gemini:
"Play my workout playlist on Spotify"
"Search for songs by The Beatles"
"Pause Spotify"
"Set a 30 minute sleep timer"
"What's currently playing?"
"Play the album 'Abbey Road' by The Beatles"
Gemini will automatically use the appropriate MCP tools to execute these commands.
API Endpoints (HTTP Bridge)
The HTTP bridge provides REST endpoints for Apple Shortcuts integration. All endpoints support optional API key authentication via:
Header:
Authorization: Bearer <api_key>Query parameter:
?apiKey=<api_key>
Play Endpoints
POST /play/playlist
POST /play/album
POST /play/track
Search
GET /search?q=query&limit=10
Returns search results for tracks, albums, and artists.
Playback Control
POST /control
GET /now-playing
Returns information about the currently playing track.
Timer
POST /timer/set
POST /timer/cancel
GET /timer/list
Returns all active timers.
Health
GET /health
Check server status and authentication state.
Apple Shortcuts Integration
Setting Up Shortcuts
Open the Shortcuts app on your iPhone/iPad
Create a new shortcut
Add a "Get Contents of URL" action
Configure:
Method: POST (or GET for search/now-playing)
URL:
http://your-server-ip:3001/play/playlistHeaders:
Content-Type: application/jsonAuthorization: Bearer <your_api_key>(if using API key)
Request Body: JSON with required parameters
Add a "Get Text from Input" action to parse the response
Add "Speak Text" action to have Siri confirm the action
Example Voice Commands
"Hey Siri, play my workout playlist"
"Hey Siri, pause Spotify"
"Hey Siri, set a 30 minute sleep timer"
"Hey Siri, what's playing on Spotify?"
"Hey Siri, play [song name] by [artist]"
Running the Server for Shortcuts
For local network access:
Find your Mac's IP address:
ifconfig | grep "inet "Use that IP in Shortcuts:
http://192.168.1.x:3001/...Ensure your Mac's firewall allows connections on port 3001
For internet access, deploy the server to a cloud provider.
Development
Project Structure
Troubleshooting
Authentication Issues
Ensure your redirect URI matches exactly in Spotify Dashboard
Check that tokens.json is created and contains valid tokens
Re-authenticate by deleting tokens.json and restarting
Device Not Found
Make sure a Spotify device is active (app open and playing)
Use the device selection in the Spotify app
Check device availability via Spotify API
Connection Issues
Verify the HTTP bridge port (default: 3001) is not blocked
Check firewall settings
Ensure the server is accessible on your network
Gemini Integration Issues
Verify the absolute path to
dist/server.jsis correctCheck environment variables are set properly
Ensure Node.js is in PATH or use full path to node
Review MCP_SETUP.md for detailed troubleshooting
License
MIT