Spotify MCP
Enables remote interaction with Spotify through the Model Context Protocol, allowing users to search tracks, control playback, and manage playlists with per-user OAuth authentication. It is hosted on Cloudflare Workers and supports remote MCP over HTTP for compatible clients.
README
Spotify MCP
Remote Spotify MCP server hosted on Cloudflare Workers.
This project exposes Spotify tools over remote MCP at /mcp and uses OAuth so each end user connects their own Spotify account. It is intended for any MCP client that supports:
- Remote MCP over HTTP
- OAuth 2.0/2.1 authorization code flow
- OAuth discovery via
/.well-known/oauth-authorization-server - Protected resource metadata via
/.well-known/oauth-protected-resource - Dynamic client registration or pre-registered OAuth clients
Poke is one supported client, but the server is no longer branded or scoped only to Poke.
What it supports
- Spotify search across tracks, albums, artists, and playlists
- Playback inspection, queue inspection, and device inspection
- Playback control: play, pause, resume, skip, queue, and volume
- Playlist reads and writes
- Album save/remove checks
- Per-user Spotify OAuth with encrypted token storage in Workers KV
Architecture
src/index.ts: Cloudflare Worker entrypoint, OAuth flow, and MCP transportsrc/spotify.ts: Spotify OAuth, token refresh, encryption, and API helperssrc/tools.ts: MCP tool definitionssrc/html.ts: landing page and authorization UI
Cloudflare setup
This Worker uses:
OAUTH_KVfor OAuth provider state and grantsSPOTIFY_TOKENSfor encrypted per-user Spotify token storage- Worker secrets for Spotify credentials and the app encryption secret
Create the KV namespaces and bind them in wrangler.jsonc, then regenerate types:
npx wrangler kv namespace create OAUTH_KV
npx wrangler kv namespace create SPOTIFY_TOKENS
npx wrangler types
Set the required secrets:
npx wrangler secret put SPOTIFY_CLIENT_ID
npx wrangler secret put SPOTIFY_CLIENT_SECRET
npx wrangler secret put APP_ENCRYPTION_SECRET
Run locally or deploy:
npx wrangler dev
npx wrangler deploy
Client setup
Configure your MCP client to use:
https://<your-worker-subdomain>.workers.dev/mcp
Clients should discover these OAuth endpoints automatically:
/.well-known/oauth-authorization-server/.well-known/oauth-protected-resource
If your client requires manual OAuth metadata, use the same base URL.
Poke example
If you are using Poke custom integrations, use the same MCP URL:
https://<your-worker-subdomain>.workers.dev/mcp
Poke should discover the OAuth metadata endpoints automatically.
Spotify dashboard setup
Your Spotify app must allow this redirect URI:
https://<your-worker-subdomain>.workers.dev/spotify/callback
Without that redirect URI configured in the Spotify Developer Dashboard, Spotify login will fail on callback.
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。
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。
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。
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。
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。
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。
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。
mcp-server-qdrant
这个仓库展示了如何为向量搜索引擎 Qdrant 创建一个 MCP (Managed Control Plane) 服务器的示例。
e2b-mcp-server
使用 MCP 通过 e2b 运行代码。