LLM Jukebox
Enables LLMs to search YouTube for music, download videos as MP3 files, and play audio with playback controls.
README
LLM Jukebox MCP Server
A Model Context Protocol (MCP) server that enables LLMs to search, download, and play YouTube music.
Features
- YouTube Music Search: Find music videos by artist, song title, album, or any search query
- Audio Download: Download and convert YouTube videos to high-quality MP3 files
- Audio Playback: Model can start and stop songs
- Async Operations: Non-blocking downloads and searches with progress tracking
- Logging: Debug-level logging to file and stderr
Installation
- Get the Files: Clone the repo or download and extract the zip
- Get FFMPEG: Make sure you have ffmpeg. In windows open a terminal and type
winget install ffmpeg, in Ubuntu or Debian dosudo apt install ffmpeg - Load a Tool Capable Model: Ensure you have a model that is trained to handle tools properly. Qwen 3 and Gemma 3 are good choices.
- Create JSON Entry: In LM Studio, click Developer mode, then Program, Tools and Integrations, the the arrow next to the Install button, and Edit mcp.json. Add the entry below under mcpServers:
"llm-jukebox": {
"command": "uv",
"args": [
"run",
"--with",
"fastmcp",
"--with",
"yt-dlp",
"--with",
"tinydb",
"--with",
"pygame",
"fastmcp",
"run",
"c:/path/to/llm-jukebox/server.py"
],
"env": {
"DOWNLOAD_PATH": "c:/path/to/downloads"
}
}
Make sure to change the paths to fit which paths the repo is in and where you want to the downloads to go.
If you have no other entries, the full JSON should look something like this:
{
"mcpServers": {
"llm-jukebox": {
"command": "uv",
"args": [
"run",
"--with",
"fastmcp",
"--with",
"yt-dlp",
"--with",
"tinydb",
"--with",
"pygame",
"fastmcp",
"run",
"c:/users/user/llm-jukebox/server.py"
],
"env": {
"DOWNLOAD_PATH": "c:/users/user/downloads"
}
}
}
}
Click on the Save button or hit Ctrl+S. If it works you should be able to set the slider to turn on llm-jukebox.
Now you can ask the LLM to grab a song for you!
Legal Considerations
This tool is intended for personal use with content you have the right to download. Users are responsible for complying with:
- YouTube's Terms of Service
- Local copyright laws
- Content creators' rights
Always respect intellectual property and consider supporting artists through official channels.
Credit
Playback tool adapted from https://github.com/Here-and-Tomorrow-LLC/audio-player-mcp (MIT Licensed)
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。