AbletonMCP
A powerful MCP server that enables Claude Desktop to control Ableton Live for automated music production, specializing in techno and industrial music generation.
README
AbletonMCP - AI Music Assistant for Ableton Live
A powerful MCP (Model Context Protocol) server that enables Claude Desktop to control Ableton Live for automated music production, specializing in techno and industrial music generation.
Features
- Natural Language Music Production: Create complete tracks from text prompts
- Genre-Specific AI: Specialized in techno and industrial music styles
- Real-time Ableton Control: Full integration with Ableton Live via OSC
- Intelligent Composition: AI-powered melody, harmony, and rhythm generation
- Professional Transport Control: Play, stop, tempo control, and project management
Prerequisites
- Python 3.10 or higher
- Ableton Live 11 Suite (or higher)
- Claude Desktop application
- At least 4GB RAM for music production
Step-by-Step Setup Guide
Step 1: Install Python Dependencies
# Navigate to the project directory
cd AbletonMCP
# Install required packages
pip install -r requirements.txt
Step 2: Install AbletonOSC Remote Script
Run the automated installer:
python install_ableton_osc.py
If the automated installer fails, install manually:
- Download AbletonOSC: https://github.com/ideoforms/AbletonOSC/archive/refs/heads/main.zip
- Extract the ZIP file
- Copy the
AbletonOSC-mainfolder to your Ableton Remote Scripts directory:- macOS:
~/Music/Ableton/User Library/Remote Scripts/ - Windows:
%USERPROFILE%/Documents/Ableton/User Library/Remote Scripts/ - Linux:
~/ableton/User Library/Remote Scripts/
- macOS:
- Rename the folder from
AbletonOSC-maintoAbletonOSC
Step 3: Configure Ableton Live 11 Suite
- Open Ableton Live 11 Suite
- Go to Live → Preferences (macOS) or Options → Preferences (Windows)
- Click on the Link/Tempo/MIDI tab
- In the Control Surface section:
- Set the first dropdown to "AbletonOSC"
- Leave Input and Output set to "None"
- Click OK to save settings
- Restart Ableton Live completely
Step 4: Test Your Setup
Test AI Composition (No Live Required):
# Test the AI composition engine offline
python examples/offline_demo.py
Test Full System (Requires Live Running):
# Run the comprehensive test suite
python test_server.py
You should see:
- ✅ Project Structure test passed
- ✅ Composition Handler test passed
- ✅ OSC Connection test passed (requires Live to be running)
- ✅ MCP Tools test passed
If OSC tests fail:
- Make sure Ableton Live 11 Suite is running
- Verify AbletonOSC is selected in Live's MIDI preferences
- Check that ports 11000/11001 aren't blocked by firewall
- The offline demo should still work perfectly!
Step 5: Start the MCP Server
python -m mcp_server.main
You should see output like:
🎵 Starting AbletonMCP Server...
✅ AbletonTools connected successfully
INFO: MCP Server listening on stdio...
Keep this terminal window open - the server needs to run continuously.
Step 6: Configure Claude Desktop
-
Quit Claude Desktop completely if it's running
-
Open your Claude Desktop configuration file:
- macOS:
~/Library/Application Support/Claude/claude_desktop_config.json - Windows:
%APPDATA%/Claude/claude_desktop_config.json
- macOS:
-
Find your Python executable path:
which python
This will show the full path to your Python installation (e.g., /Users/yourname/anaconda3/bin/python or /opt/homebrew/bin/python3).
- Create or edit the claude_desktop_config.json file with this content:
{
"mcpServers": {
"ableton-mcp": {
"command": "/full/path/to/your/python",
"args": ["-m", "mcp_server.main"],
"cwd": "/full/path/to/your/AbletonMCP",
"env": {
"PYTHONPATH": "/full/path/to/your/AbletonMCP"
}
}
}
}
Important:
- Replace
/full/path/to/your/pythonwith the result fromwhich pythoncommand - Replace
/full/path/to/your/AbletonMCPwith the actual full path to your AbletonMCP project directory (appears twice) - The
PYTHONPATHenvironment variable is crucial for Python to find the mcp_server module - Use absolute paths, not relative paths like
pythonor~/
- Save the file and restart Claude Desktop
Step 7: Verify Connection in Claude Desktop
- Open Claude Desktop
- Start a new conversation
- You should see "AbletonMCP" listed in the server connections at the bottom
- If it shows as connected, you're ready to make music!
Example Prompts - Test All Features
Basic Transport Control
Try these prompts in Claude Desktop:
1. Test Connection
Ping Ableton Live to test the connection
2. Tempo Control
Set the tempo to 132 BPM
3. Transport Control
Start playback
Stop playback after 8 seconds
Track Management
4. Create Basic Tracks
Create a MIDI track called "Kick Drum"
Create a MIDI track called "Bass Synth"
Create an audio track called "Percussion"
Create a return track called "Industrial Reverb"
AI Music Generation
5. Generate Chord Progressions
Generate a chord progression in A minor for industrial techno, 8 bars long
Generate a chord progression in D minor for minimal techno, 4 bars
6. Create Drum Patterns
Create an industrial drum pattern for 4 bars
7. Complete Song Generation
Create a 64-bar industrial techno song at 132 BPM in A minor
Advanced Features
8. Project Information
Get information about the current Ableton Live session
9. Tempo Variations
Set tempo to 128 BPM for underground techno
Set tempo to 140 BPM for hard techno
Complex Compositions
10. Full Production Workflow
Create a complete techno track: Set tempo to 132 BPM, create kick, bass, lead, and percussion tracks, then generate an industrial techno song structure for 64 bars in A minor. Add atmospheric elements and start playback to preview.
Expected Results in Ableton Live
When you run these prompts, you should see:
- Transport changes: BPM adjustments, play/stop states
- New tracks appearing: Named tracks in the Session View
- MIDI clips generated: Containing chord progressions and patterns
- Console feedback: Success messages in both Claude and the MCP server terminal
Troubleshooting
"Could not connect to Ableton Live"
- ✅ Ensure Ableton Live is running
- ✅ Verify AbletonOSC is installed and enabled
- ✅ Check Live's MIDI preferences show "AbletonOSC" selected
- ✅ Restart both Live and the MCP server
"MCP Server not found in Claude Desktop"
- ✅ Check the config.json path is correct for your OS
- ✅ Verify the "cwd" path points to your AbletonMCP directory
- ✅ Make sure to restart Claude Desktop after config changes
- ✅ Check Claude Desktop's console for error messages
"Permission denied" or import errors
- ✅ Run
pip install -r requirements.txt - ✅ Use Python 3.10 or higher
- ✅ Consider using a virtual environment
Music generation not working as expected
- ✅ This is Phase 1 - basic functionality
- ✅ Try different musical keys (Am, Dm, Em, Gm)
- ✅ Experiment with different BPM ranges (120-150)
- ✅ Use genre keywords: "industrial", "minimal", "underground"
Next Steps
Once everything is working:
- Experiment with different musical styles and BPMs
- Combine prompts for complex arrangements
- Use Live's tools to further refine the AI-generated content
- Save your projects in Ableton Live as usual
What's Working in Phase 1
✅ Transport Control - Play/stop, tempo setting
✅ Track Creation - MIDI, audio, and return tracks
✅ AI Chord Progressions - Genre-specific harmony generation
✅ Basic Song Structures - Multi-section arrangements
✅ OSC Communication - Real-time Live control
✅ Music Theory Integration - Coherent musical output
Project Structure
mcp_server/- MCP server implementationableton_control/- OSC and Ableton Live API wrappermusic_ai/- AI composition and generation engineaudio_processing/- Audio analysis and processing toolsknowledge_base/- Music theory and genre-specific datasamples/- Audio content librarytests/- Test suite
Development
Install development dependencies:
pip install -r requirements.txt
Run tests:
pytest
Format code:
black . && isort .
License
MIT License - see LICENSE file for details.
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。