Spotify MCP Server
Enables AI agents to search Spotify, inspect playback state, manage devices, and control music through the official Spotify Web API using OAuth authentication.
README
Spotify MCP Server
Spotify MCP Server for AI agents: a Streamable HTTP Model Context Protocol server that lets agents search Spotify, inspect playback state, manage devices and control music through the official Spotify Web API.
Status
Early design phase. The first target is a local single-user MCP server with Spotify OAuth, Docker support and a small stable tool surface.
Goals
- Expose Spotify Web API capabilities as MCP tools.
- Use official OAuth instead of browser automation.
- Support local personal agents first.
- Keep the project reusable by different MCP clients.
- Provide a future path for multi-user agent backends.
Initial Scope
The first stable version will focus on:
- Streamable HTTP MCP transport.
- Local single-user Spotify OAuth.
- Automatic access token refresh.
- Profile, search, devices and playback state tools.
- Basic playback control tools, including playback transfer between Spotify Connect devices.
- Docker-based runtime.
Playlist write operations and external token mode are planned after the local single-user flow is stable.
Requirements
- Spotify account.
- Spotify Developer application.
- Spotify Premium for playback control endpoints.
- Node.js runtime for local development.
- Docker for containerized usage.
Default Ports
- MCP HTTP server:
11070 - OAuth callback:
http://127.0.0.1:11070/auth/callback
Local Development
Install dependencies:
npm install
Run type checks:
npm run typecheck
Build:
npm run build
Start the compiled server:
npm start
dev and start load .env automatically.
Healthcheck:
curl http://localhost:11070/health
Auth status:
curl http://localhost:11070/auth/status
MCP endpoint:
http://localhost:11070/mcp
The initial tool surface exposes:
spotify_ping: validates MCP transport without calling Spotify.spotify_get_profile: calls Spotify Web API/meusing the local OAuth token.spotify_search: searches tracks, artists, albums and playlists.spotify_get_devices: lists Spotify Connect devices.spotify_get_playback_state: reads current playback state.spotify_get_current_track: reads the current track or episode.spotify_play: starts or resumes playback.spotify_play_search: searches a track and plays the best match.spotify_transfer_playback: transfers playback to another Spotify Connect device.spotify_pause: pauses playback.spotify_next: skips to the next track.spotify_previous: skips to the previous track.spotify_set_volume: sets playback volume.
Docker
Build and run with the example compose file:
cp .env.example .env
docker compose -f docker-compose.example.yml up --build
Local Spotify OAuth
Create a Spotify Developer application with this redirect URI:
http://127.0.0.1:11070/auth/callback
Then configure .env and open:
http://localhost:11070/auth/login
See Local Spotify OAuth Setup.
Access tokens are refreshed automatically when they are expired or close to expiration.
Documentation
Security
Do not commit secrets or tokens. Use .env for Spotify application credentials
and keep local token files out of git.
The project should request only the scopes required by its enabled tools. New write permissions should be introduced deliberately and documented before they are implemented.
License
License pending.
推荐服务器
Baidu Map
百度地图核心API现已全面兼容MCP协议,是国内首家兼容MCP协议的地图服务商。
Playwright MCP Server
一个模型上下文协议服务器,它使大型语言模型能够通过结构化的可访问性快照与网页进行交互,而无需视觉模型或屏幕截图。
Magic Component Platform (MCP)
一个由人工智能驱动的工具,可以从自然语言描述生成现代化的用户界面组件,并与流行的集成开发环境(IDE)集成,从而简化用户界面开发流程。
Audiense Insights MCP Server
通过模型上下文协议启用与 Audiense Insights 账户的交互,从而促进营销洞察和受众数据的提取和分析,包括人口统计信息、行为和影响者互动。
VeyraX
一个单一的 MCP 工具,连接你所有喜爱的工具:Gmail、日历以及其他 40 多个工具。
graphlit-mcp-server
模型上下文协议 (MCP) 服务器实现了 MCP 客户端与 Graphlit 服务之间的集成。 除了网络爬取之外,还可以将任何内容(从 Slack 到 Gmail 再到播客订阅源)导入到 Graphlit 项目中,然后从 MCP 客户端检索相关内容。
Kagi MCP Server
一个 MCP 服务器,集成了 Kagi 搜索功能和 Claude AI,使 Claude 能够在回答需要最新信息的问题时执行实时网络搜索。
e2b-mcp-server
使用 MCP 通过 e2b 运行代码。
Neon MCP Server
用于与 Neon 管理 API 和数据库交互的 MCP 服务器
Exa MCP Server
模型上下文协议(MCP)服务器允许像 Claude 这样的 AI 助手使用 Exa AI 搜索 API 进行网络搜索。这种设置允许 AI 模型以安全和受控的方式获取实时的网络信息。