mcp-score
AI-powered music notation server that lets you create and edit scores using natural language, integrating with MuseScore for live manipulation.
README
mcp-score
AI-powered music notation. Describe what you want in plain English -- get a publication-ready score in MuseScore.
Quick demo
"Create a big band chart -- 32-bar AABA form, key of Bb, slow blues at 66 BPM, with rhythm changes and rehearsal marks at each section."
Claude writes a complete music21 script, executes it, and hands you a MusicXML file ready to open in MuseScore, Dorico, Sibelius, or any notation app.
With the MuseScore plugin running, you can go further:
"Read the melody in bars 9-16 and arrange it as a trombone soli following the chord progression."
Claude reads the live score, applies musical intelligence, and writes the arrangement back -- all through natural language.
Installation
pip install mcp-score
Or with uv:
uv tool install mcp-score
Then set up the components you need:
# Install the score generation skill (for Claude Code)
mcp-score install-skill
# Install the MuseScore plugin (for live score manipulation)
mcp-score install-plugin
# Or install both at once
mcp-score install
Components
mcp-score has three components that work together:
MCP server
A Python MCP server with 18 tools for live score manipulation across MuseScore, Dorico, and Sibelius: connect/disconnect, read passages, add chords, set barlines, transpose, and more. Runs via mcp-score serve (or just mcp-score).
Score generation skill
A Claude Code skill that teaches Claude to write music21 Python scripts for score generation. Installed to ~/.claude/skills/score-generate/. This handles the "create a score from scratch" use case -- no MCP round-trips needed.
MuseScore plugin
A QML plugin that runs a WebSocket server inside MuseScore 4, enabling the MCP server to read from and write to the active score in real time. Supports 19 commands including navigation, note input, chord symbols, rehearsal marks, barlines, key/time signatures, tempo, transposition, and undo.
Configuration
Claude Code
Add the MCP server to your project or global settings:
claude mcp add mcp-score -- mcp-score serve
The score generation skill is installed separately with mcp-score install-skill and activates automatically when you ask Claude to create a score.
Claude Desktop
Add to your claude_desktop_config.json:
{
"mcpServers": {
"mcp-score": {
"command": "mcp-score",
"args": ["serve"]
}
}
}
MuseScore plugin setup
After mcp-score install-plugin:
- Open MuseScore 4
- Go to Plugins > Plugin Manager
- Enable MCP Score Bridge
- The plugin starts a WebSocket server on port 8765
See MuseScore plugin docs for details.
CLI reference
mcp-score serve Run the MCP server (default)
mcp-score run <script> Run a Python script with music21 available
mcp-score install Install skill and MuseScore plugin
mcp-score install-skill Install the score-generate skill to ~/.claude/skills/
mcp-score install-plugin Install the QML plugin to MuseScore's Plugins directory
mcp-score help Show help
Documentation
| Document | Description |
|---|---|
| Getting started | Set up mcp-score and generate your first score |
| Architecture | System design and key decisions |
| Tool reference | Complete list of MCP tools |
| MuseScore plugin | Plugin installation and WebSocket protocol |
Contributing
See CONTRIBUTING.md for development setup, code style, and PR process.
Author
Thomas Skovlund Hansen — skovlund.dev · thomas@skovlund.dev
License
Recommended Servers
playwright-mcp
A Model Context Protocol server that enables LLMs to interact with web pages through structured accessibility snapshots without requiring vision models or screenshots.
Magic Component Platform (MCP)
An AI-powered tool that generates modern UI components from natural language descriptions, integrating with popular IDEs to streamline UI development workflow.
Audiense Insights MCP Server
Enables interaction with Audiense Insights accounts via the Model Context Protocol, facilitating the extraction and analysis of marketing insights and audience data including demographics, behavior, and influencer engagement.
VeyraX MCP
Single MCP tool to connect all your favorite tools: Gmail, Calendar and 40 more.
graphlit-mcp-server
The Model Context Protocol (MCP) Server enables integration between MCP clients and the Graphlit service. Ingest anything from Slack to Gmail to podcast feeds, in addition to web crawling, into a Graphlit project - and then retrieve relevant contents from the MCP client.
Kagi MCP Server
An MCP server that integrates Kagi search capabilities with Claude AI, enabling Claude to perform real-time web searches when answering questions that require up-to-date information.
E2B
Using MCP to run code via e2b.
Neon Database
MCP server for interacting with Neon Management API and databases
Exa Search
A Model Context Protocol (MCP) server lets AI assistants like Claude use the Exa AI Search API for web searches. This setup allows AI models to get real-time web information in a safe and controlled way.
Qdrant Server
This repository is an example of how to create a MCP server for Qdrant, a vector search engine.