Spotify MCP Server

Streamable HTTP MCP server for Spotify — search music, control playback, manage playlists and saved songs.

Author: overment

[!WARNING] This warning applies only to the HTTP transport and OAuth wrapper included for convenience. They are intended for personal/local use and are not production‑hardened.

The MCP tools and schemas themselves are implemented with strong validation, slim outputs, clear error handling, and other best practices.

If you plan to deploy remotely, harden the OAuth/HTTP layer: proper token validation, secure storage, TLS termination, strict CORS/origin checks, rate limiting, audit logging, and compliance with Spotify's terms.

Motivation

At first glance, a "Spotify MCP" may seem unnecessary—pressing play or skipping a song is often faster by hand. It becomes genuinely useful when you don't know the exact title (e.g., "soundtrack from [movie title]"), when you want to "create and play a playlist that matches my mood", or when you're using voice. This MCP lets an LLM handle the fuzzy intent → search → selection → control loop, and it returns clear confirmations of what happened. It works well with voice interfaces and can be connected to agents/workflows for smart‑home automations.

Demo

Alice — a desktop AI assistant

Claude Desktop

Features

• ✅ Search — Find tracks, albums, artists, playlists
• ✅ Player Control — Play, pause, skip, seek, volume, shuffle, repeat, queue
• ✅ Device Transfer — Move playback between devices
• ✅ Playlists — Create, edit, add/remove tracks, reorder
• ✅ Library — Save/remove tracks, check if saved
• ✅ OAuth 2.1 — Secure PKCE flow with RS token mapping
• ✅ Dual Runtime — Node.js/Bun or Cloudflare Workers
• ✅ Production Ready — Encrypted token storage, rate limiting, multi-user support

Design Principles

• LLM-friendly: Tools don't mirror Spotify's API 1:1 — interfaces are simplified and unified
• Batch-first: Operations use arrays (queries[], operations[]) to minimize tool calls
• Clear feedback: Every response includes human-readable _msg with what succeeded/failed
• Best-effort verification: Player control verifies device, context, and current track

Quick Start

1. Install

cd spotify-mcp
bun install

2. Configure

cp .env.example .env

Edit .env:

PORT=3000
AUTH_ENABLED=true

# From https://developer.spotify.com/dashboard
SPOTIFY_CLIENT_ID=your_client_id
SPOTIFY_CLIENT_SECRET=your_client_secret

# OAuth
OAUTH_SCOPES=playlist-read-private playlist-read-collaborative playlist-modify-public playlist-modify-private user-read-playback-state user-modify-playback-state user-read-currently-playing user-library-read user-library-modify
OAUTH_REDIRECT_URI=alice://oauth/callback
OAUTH_REDIRECT_ALLOWLIST=alice://oauth/callback

3. Configure Spotify Dashboard

Add redirect URIs in Spotify Developer Dashboard:

http://127.0.0.1:3001/oauth/callback
alice://oauth/callback

4. Run

bun dev
# MCP: http://127.0.0.1:3000/mcp
# OAuth: http://127.0.0.1:3001

Server Instructions (What the Model Sees)

Use these tools to find music, get the current player status, control and transfer playback, and manage playlists and saved songs.

Tools
- search_catalog: Find songs, artists, albums, or playlists
- player_status: Read current player, available devices, queue, and current track
- spotify_control: Batch control playback (play, pause, next, previous, seek, volume, shuffle, repeat, transfer, queue)
- spotify_playlist: Manage playlists (list, get, items, create, update, add/remov

... [View full README on GitHub](https://github.com/iceener/spotify-streamable-mcp-server#readme)