About ableton-copilot-mcp
ableton-copilot-mcp is an MCP server built on ableton-js that lets AI assistants control Ableton Live's Arrangement View in real time. It covers songs, tracks, MIDI editing and audio recording, and is aimed at music producers who want help with repetitive production tasks. It is a third-party integration and not made by Ableton.
Key features
- 01Reads song information and track lists
- 02Creates, duplicates and deletes tracks
- 03Adds, deletes and replaces MIDI notes
- 04Loads devices and modifies their parameters
- 05Operation history with note rollback
Use cases
- 01Batch-editing MIDI notes such as humanizing or merging notes
- 02Recording a track's content to an audio track
- 03Adjusting track properties and device parameters with AI help
MCP support
This project is an MCP server that uses ableton-js to expose tools for controlling Ableton Live songs, tracks, clips and devices to MCP clients.
Configuration
Local server config is the same for every client. Copy it into your client’s MCP config file.
Tools
35Saved tool list · updated 3w ago
| Parameter | Type | Required | Description |
|---|---|---|---|
| type | enum | Yes | [string] resources typeaudio_effects | instruments | plugins | samples | drums | midi_effects | sounds |
Details
From the project READMEableton-copilot-mcp
An MCP (Model Context Protocol) server built on ableton-js for real-time interaction and control with Ableton Live's Arrangement View, dedicated to assisting music producers in their music production.
🎯 Note
As a music producer, I have some understanding of using Ableton for music creation. During the creative process, we often need to handle various tedious operations, such as humanizing note properties, merging notes, recording one track to another audio track, etc. Previously, we could only rely on the functions provided by the host DAW for these operations. However, with the emergence of large language models and MCP, we now have the possibility to let AI help us with automation. Although it's still not realistic for AI to directly generate notes or create a complete song (it can't generate a good-sounding song), having AI assist us with auxiliary operations presents a new possibility.
🎥 Demo
This is a simple demonstration video of MIDI operations.
https://github.com/user-attachments/assets/8d635c4c-a1f2-44be-93ae-062038d14c71
🚀 Features
🎵 Song Control
- Get basic song information (root note, scale name, tempo, song length, etc.)
- Get a list of all tracks
- Create MIDI, audio, and return tracks
- Delete and duplicate tracks
🎹 Track Management
- Get all clips in a track
- Create empty MIDI clips in the arrangement view tracks
- Create audio clips in tracks based on provided sample file paths
- Set track properties (mute, color, name, arm, solo, etc.)
- Duplicate MIDI clips to specified tracks
🎼 Clip Operations
- Get clips in the piano roll view
- Get and manage all notes in a clip
- Add, delete, and replace notes in clips
- Set clip properties (name, color, looping, loop_start, loop_end, etc.)
🎧 Audio Operations
- Supports recording track content based on time range
🔌 Device Management
- Load audio effects, instruments and plugins
- Modify device parameters
- Browse available devices in the library
📝 State Management
- Operation history tracking and detailed logging
- Support for operation rollback (currently only supports note operations)
- Browse and restore previous states
📝 To-Do List
- Support automatic envelope adjustment
- Support more snapshot types (track properties, device parameters, etc.)
⚠️ Warning
- Direct manipulation of MIDI clips by AI may result in the loss of original notes and cannot be undone with Ctrl + Z. Please operate with caution. If needed, you can ask the AI to help you roll back note operations.
📥 Installation
Prerequisites
- Node.js environment: Ensure Node.js is installed (v20+ recommended) and the
npxcommand is available🔗 Download: Node.js official website
Installation Steps
1. Install AbletonJS MIDI Remote Scripts
Choose ONE of the following three methods to install MIDI Remote Scripts:
-
Method 1: One-line installation (Recommended)
npx @xiaolaa2/ableton-copilot-mcp --install-scripts # or use the short form npx @xiaolaa2/ableton-copilot-mcp -is -
Method 2: Tool-assisted installation
- First, start ableton-copilot-mcp (see Usage section below)
- Let AI assistant or manually call the MCP tool
init_ableton_jsto complete the installation automatically
-
Method 3: Manual installation
- Create a folder named "Remote Scripts" in your Ableton User Library:
- Windows:
C:\Users\[username]\Documents\Ableton\User Library\Remote Scripts - Mac:
/Users/[username]/Music/Ableton/User Library/Remote Scripts
- Windows:
- Download MIDI Remote Scripts from the ableton-js project
- Copy the downloaded
midi-scriptfolder to the location above - Rename it to
AbletonJS
- Create a folder named "Remote Scripts" in your Ableton User Library:
🔧 Usage
Quick Start
-
Launch Ableton Live
-
Enable AbletonJS Control Surface
- Open Ableton Live Preferences:
Preferences→Link/MIDI - In the
MIDItab, locate theControl Surfacesection - Select
AbletonJSfrom the dropdown menu
Figure 1: Enabling AbletonJS Control Surface in Ableton Live Configuration
- Open Ableton Live Preferences:
-
Connect using an MCP client
The following clients can be used as MCP endpoints:
- Cursor
- Cherry Studio
- Claude Desktop
- Other MCP clients
Client Configuration
Cursor Configuration Example
Add the following configuration to your Cursor settings:
"ableton-js-mcp": {
"command": "npx",
"args": [
"-y",
"@xiaolaa2/ableton-copilot-mcp"
]
}
For the latest version:
"ableton-js-mcp": {
"command": "npx",
"args": [
"-y",
"@xiaolaa2/ableton-copilot-mcp@latest"
]
}
Custom Storage Path (Optional)
You can specify a custom data storage location using the BASE_PATH environment variable:
"ableton-js-mcp": {
"command": "npx",
"args": [
"-y",
"@xiaolaa2/ableton-copilot-mcp"
],
"env": {
"BASE_PATH": "D:\\ableton_copilot_mcp"
}
}
💡 Tip: The storage path is used to save log files, operation history, and state snapshots
✅ Compatibility Testing
Compatibility primarily depends on the version support of the ableton-js library.
| Ableton Live Version | Test Status |
|---|---|
| 12.1.10 | ✅ Tested and working |
| 11.x | ⚠️ Not tested yet |
| 10.x | ⚠️ Not tested yet |
🤝 Contributing
Issues and contributions are welcome. Please submit issues or suggestions through GitHub Issues.
📄 License
This project is licensed under the MIT License.
⚠️ Disclaimer
This is a third-party integration and not made by Ableton.
Comments
New to MCP?
From what the protocol is to installing your first server, these answers cover most first questions.
An open protocol that defines how AI applications connect to external tools and data in a uniform way. Clients talk to the model; MCP servers provide tools, resources and prompt templates. They exchange standard messages, so one server can be reused by many clients.