Tubist MCP Server

Tubist MCP Server

Enables AI assistants to control the Tubist macOS menu bar player to play YouTube videos using natural language commands, allowing uninterrupted background audio.

Category
访问服务器

README

Tubist MCP Server

Remember the last time when you closed a tab or quit your browser and YouTube audio abruptly stopped...? Bet you were angry.

Tubist exists for that not to happen. It's a lightweight macOS menu bar player that keeps YouTube going independently of your browser.

Tubist is available for free on the Mac App Store.

This MCP server lets Claude and other AI assistans control Tubist when asked to do so from the assistant's macOS app.

Just ask naturally:

  • play the Billie Eilish Tiny Desk concert
  • find me some jazz on YouTube for focus and play it in Tubist
  • look for an orchestral rendering of Super Mario music on YouTube and play it in Tubist

Features

  • 🎬 Play any YouTube video with a simple request... (unless it longer exists 😞)
  • 🎵 Ideal for background audio (concerts, podcasts, music)
  • 🤖 Control via natural language through AI assistants
  • ⚡ Instant playback - just ask and it plays without any UI interruptions

Prerequisites / minimum requirements

Installation

  1. Clone this repository:

    git clone https://github.com/dkasaj/tubist-mcp-server.git
    cd tubist-mcp-server
    
  2. Install dependencies:

    npm install
    
  3. Configure Claude Desktop:

    Edit your Claude Desktop config file:

    # On macOS
    nano ~/Library/Application\ Support/Claude/claude_desktop_config.json
    

    Add the Tubist MCP server, use the path where you installed it:

    {
      "mcpServers": {
        "Tubist - menu bar player for YouTube": {
     	 "command": "/opt/homebrew/bin/node",
     	 "args": ["/path/to/tubist-mcp-server/index.js"]
        }
      }
    }
    

    Important: Replace /path/to/tubist-mcp/index.js with the actual path where you cloned this repo.

    Find your Node.js path:

    which node
    

    Use this path instead of /opt/homebrew/bin/node if different.

  4. Restart Claude Desktop completely (quit and reopen, don't just close the window)

Usage

Once configured, you can ask Claude to play YouTube videos naturally:

Examples:

  • "Play the Billie Eilish Tiny Desk concert"
  • "Find me some jazz on YouTube for focus and play it in Tubist"
  • "Look for an orchestral rendering of Super Mario music on YouTube and play it in Tubist"

How It Works

  1. You make a request to Claude
  2. Claude uses the MCP protocol to call this server
  3. The server opens the YouTube video in Tubist using the tubist://play? URL scheme
  4. Video starts playing automatically

Troubleshooting

"MCP server disconnected" error

  • Check that the path in claude_desktop_config.json is correct
  • Verify Node.js path with which node
  • Make sure you restarted Claude Desktop completely

Video doesn't play

  • Ensure Tubist is installed and can be launched
  • Try opening Tubist manually first to verify it works
  • Check that the YouTube URL is valid. Claude for instance sometimes uses a link to a video that no longer exists.

Tool doesn't appear in Claude

  • Confirm the config file is valid JSON (use a JSON validator)
  • Check for log Claude Desktop logs: ~/Library/Logs/Claude/
  • Make sure you saved the config file and restarted Claude

Technical Details

This MCP server uses the Server class with setRequestHandler pattern (rather than McpServer.tool()) due to a known bug in MCP SDK 1.25.x where arguments aren't passed correctly to tool handlers.

Tool definition:

  • Name: play
  • Parameter: url (YouTube video URL)
  • Action: Opens video in Tubist via tubist://play?[url] URL scheme

Contributing

Contributions welcome! Please feel free to submit a Pull Request.

License

MIT © Daniel Kašaj

Links


Made with ❤️ for background YouTube listening

推荐服务器

Baidu Map

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。

官方
精选
JavaScript
Playwright MCP Server

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。

官方
精选
TypeScript
Magic Component Platform (MCP)

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。

官方
精选
本地
TypeScript
Audiense Insights MCP Server

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。

官方
精选
本地
TypeScript
VeyraX

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。

官方
精选
本地
graphlit-mcp-server

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。

官方
精选
TypeScript
Kagi MCP Server

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。

官方
精选
Python
e2b-mcp-server

e2b-mcp-server

使用 MCP 通过 e2b 运行代码。

官方
精选
Neon MCP Server

Neon MCP Server

用于与 Neon 管理 API 和数据库交互的 MCP 服务器

官方
精选
Exa MCP Server

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。

官方
精选