Ableton MCP Extended
An independently maintained MCP server that adds deeper session editing, mixing, MIDI surgery, and bounce workflows for Ableton Live, with enhanced privacy and security.
README
Ableton MCP Extended
An independently maintained, privacy-focused derivative of ahujasid/ableton-mcp that adds deeper session editing, feedback, mixing, MIDI surgery, and bounce workflows for Ableton Live.
This project is not affiliated with or endorsed by Ableton. It remains in private staging while the extended commands are tested against real Live sets.
Why this derivative exists
The upstream project provides the MCP-to-Ableton bridge. This derivative keeps that architecture and adds 31 MCP tools:
- Session structure: create audio and return tracks, delete tracks, create and name scenes, duplicate clips between scenes, and inspect scene names.
- Clip editing: delete clips, clear notes, read notes, update individual notes, remove notes by pitch/time range, set loop markers, and quantize clips.
- Instruments and devices: load samples onto Drum Rack pads, inspect pad names, inspect device parameters, and set device parameters.
- Mixing and feedback: set volume, pan, mute, solo, arm, monitoring, color, and send levels; inspect return tracks and output meters; undo and redo.
- Rendering: bounce a Session View MIDI clip to audio in real time.
It also changes the privacy posture:
- No telemetry code or Supabase dependency.
- The Ableton Remote Script listens on
127.0.0.1by default, not every network interface. - MCP request and response sizes are bounded.
- Command parameters, prompts, paths, and raw responses are not written to normal MCP logs.
See UPSTREAM.md for provenance and CHANGELOG.md for the maintained change summary.
Requirements
- Ableton Live 11 or newer; Live 12 is recommended.
- Python 3.10 or newer for the MCP server.
- uv.
Some note-editing and clip-duplication operations depend on APIs introduced in newer Live versions. Save a copy of important sets before testing destructive commands.
Install for development
git clone https://github.com/RichieMiller-HU/ableton-mcp-extended.git
cd ableton-mcp-extended
uv sync --extra dev
Copy AbletonMCP_Remote_Script into Ableton's User Remote Scripts directory,
restart Live, and select AbletonMCP as a Control Surface.
During private staging, configure an MCP client to run the local checkout:
{
"mcpServers": {
"AbletonMCP-Extended": {
"command": "uv",
"args": [
"--directory",
"/absolute/path/to/ableton-mcp-extended",
"run",
"ableton-mcp-extended"
]
}
}
}
The path above is an example. Never commit a personal absolute path or MCP client configuration containing credentials.
Only run one Ableton MCP server at a time.
Security model
The Remote Script has no authentication. Loopback binding prevents direct
connections from other computers, but any process running as a local user may
attempt to connect to port 9877 and issue Ableton commands. Do not change the
bind address to 0.0.0.0 or expose the port through forwarding or tunneling.
See SECURITY.md for reporting and deployment guidance.
Development checks
uv run --extra dev pytest
uv run --extra dev detect-secrets scan --all-files --exclude-files '^\.git/'
uv build
Live API behavior still requires an integration pass inside Ableton; the unit suite covers the Python bridge and publication/privacy invariants.
License and attribution
MIT licensed. The original copyright and permission notice are preserved in LICENSE. Modifications are documented rather than presented as original upstream work.
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。