MuseScore MCP Server

MuseScore MCP Server

A Model Context Protocol server that provides programmatic control over MuseScore through a WebSocket-based plugin system, allowing AI assistants to compose music, add lyrics, navigate scores, and control MuseScore directly.

Category
访问服务器

README

MuseScore MCP Server

A Model Context Protocol (MCP) server that provides programmatic control over MuseScore through a WebSocket-based plugin system. This allows AI assistants like Claude to compose music, add lyrics, navigate scores, and control MuseScore directly.

Demo GIF

Prerequisites

  • MuseScore 3.x or 4.x
  • Python 3.8+
  • Claude Desktop or compatible MCP client

Setup

1. Install the MuseScore Plugin

First, save the QML plugin code to your MuseScore plugins directory:

macOS: ~/Documents/MuseScore4/Plugins/musescore-mcp-websocket.qml Windows: %USERPROFILE%\Documents\MuseScore4\Plugins\musescore-mcp-websocket.qml Linux: ~/Documents/MuseScore4/Plugins/musescore-mcp-websocket.qml

2. Enable the Plugin in MuseScore

  1. Open MuseScore
  2. Go to Plugins → Plugin Manager
  3. Find "MuseScore API Server" and check the box to enable it
  4. Click OK

3. Setup Python Environment

git clone <your-repo>
cd mcp-agents-demo
python -m venv .venv
source .venv/bin/activate  # On Windows: .venv\Scripts\activate
pip install fastmcp websockets

4. Configure Claude Desktop

Add to your Claude Desktop configuration file:

macOS: ~/Library/Application Support/Claude/claude_desktop_config.json Windows: %APPDATA%\Claude\claude_desktop_config.json

{
  "mcpServers": {
    "musescore": {
      "command": "/path/to/your/project/.venv/bin/python",
      "args": [
        "/path/to/your/project/server.py"
      ]
    }
  }
}

Note: Update the paths to match your actual project location.

Running the System

Order of Operations (Important!)

  1. Start MuseScore first with a score open
  2. Run the MuseScore plugin: Go to Plugins → MuseScore API Server
    • You should see console output: "Starting MuseScore API Server on port 8765"
  3. Then start the Python MCP server or restart Claude Desktop

[insert screenshot of different functionality, harmonisation, melodywriting, as zoomed in GIFs]

Development and Testing

For development, use the MCP development tools:

# Install MCP dev tools
pip install mcp

# Test your server
mcp dev server.py

# Check connection status
mcp dev server.py --inspect

Viewing Console Output

To see MuseScore plugin console output, run MuseScore from terminal:

macOS:

/Applications/MuseScore\ 4.app/Contents/MacOS/mscore

Windows:

cd "C:\Program Files\MuseScore 4\bin"
MuseScore.exe

Linux:

musescore4

Features

This MCP server provides comprehensive MuseScore control:

Navigation & Cursor Control

  • get_cursor_info() - Get current cursor position and selection info
  • go_to_measure(measure) - Navigate to specific measure
  • go_to_beginning_of_score() / go_to_final_measure() - Navigate to start/end
  • next_element() / prev_element() - Move cursor element by element
  • next_staff() / prev_staff() - Move between staves
  • select_current_measure() - Select entire current measure

Note & Rest Creation

  • add_note(pitch, duration, advance_cursor_after_action) - Add notes with MIDI pitch
  • add_rest(duration, advance_cursor_after_action) - Add rests
  • add_tuplet(duration, ratio, advance_cursor_after_action) - Add tuplets (triplets, etc.)

Measure Management

  • insert_measure() - Insert measure at current position
  • append_measure(count) - Add measures to end of score
  • delete_selection(measure) - Delete current selection or specific measure

Lyrics & Text

  • add_lyrics_to_current_note(text) - Add lyrics to current note
  • add_lyrics(lyrics_list) - Batch add lyrics to multiple notes
  • set_title(title) - Set score title

Score Information

  • get_score() - Get complete score analysis and structure
  • ping_musescore() - Test connection to MuseScore
  • connect_to_musescore() - Establish WebSocket connection

Utilities

  • undo() - Undo last action
  • set_time_signature(numerator, denominator) - Change time signature
  • processSequence(sequence) - Execute multiple commands in batch

Sample Music

Check out the /examples folder for sample MuseScore files demonstrating various musical styles:

  • Asian Instrumental - Traditional Asian-inspired instrumental piece
  • String Quartet - Classical string quartet arrangement

Each example includes:

  • .mscz - MuseScore file (editable)
  • .pdf - Sheet music
  • .mp3 - Audio preview

Usage Examples

Creating a Simple Melody

# Set up the score
await set_title("My First Song")
await go_to_beginning_of_score()

# Add notes (MIDI pitch: 60=C, 62=D, 64=E, etc.)
await add_note(60, {"numerator": 1, "denominator": 4}, True)  # Quarter note C
await add_note(64, {"numerator": 1, "denominator": 4}, True)  # Quarter note E
await add_note(67, {"numerator": 1, "denominator": 4}, True)  # Quarter note G
await add_note(72, {"numerator": 1, "denominator": 2}, True)  # Half note C

# Add lyrics
await go_to_beginning_of_score()
await add_lyrics_to_current_note("Do")
await next_element()
await add_lyrics_to_current_note("Mi")
await next_element()
await add_lyrics_to_current_note("Sol")
await next_element()
await add_lyrics_to_current_note("Do")

Batch Operations

# Add multiple lyrics at once
await add_lyrics(["Twin-", "kle", "twin-", "kle", "lit-", "tle", "star"])

# Use sequence processing for complex operations
sequence = [
    {"action": "goToBeginningOfScore", "params": {}},
    {"action": "addNote", "params": {"pitch": 60, "duration": {"numerator": 1, "denominator": 4}, "advanceCursorAfterAction": True}},
    {"action": "addNote", "params": {"pitch": 64, "duration": {"numerator": 1, "denominator": 4}, "advanceCursorAfterAction": True}},
    {"action": "addRest", "params": {"duration": {"numerator": 1, "denominator": 4}, "advanceCursorAfterAction": True}}
]
await processSequence(sequence)

Troubleshooting

Connection Issues

  • "Not connected to MuseScore":
    • Ensure MuseScore is running with a score open
    • Run the MuseScore plugin (Plugins → MuseScore API Server)
    • Check that port 8765 isn't blocked by firewall

Plugin Issues

  • Plugin not appearing: Check the .qml file is in the correct plugins directory
  • Plugin won't enable: Restart MuseScore after placing the plugin file
  • No console output: Run MuseScore from terminal to see debug messages

Python Server Issues

  • "No server object found": The server object must be named mcp, server, or app at module level
  • WebSocket errors: Make sure MuseScore plugin is running before starting Python server
  • Connection timeout: The MuseScore plugin must be actively running, not just enabled

API Limitations

  • Lyrics: Only first verse supported in MuseScore 3.x plugin API
  • Title setting: Uses multiple fallback methods due to frame access limitations
  • Selection persistence: Some operations may affect current selection

File Structure

mcp-agents-demo/
├── .venv/
├── server.py                           # Python MCP server entry point
├── musescore-mcp-websocket.qml         # MuseScore plugin
├── requirements.txt
├── README.md
└── src/                                # Source code modules
    ├── __init__.py
    ├── client/                         # WebSocket client functionality
    │   ├── __init__.py
    │   └── websocket_client.py
    ├── tools/                          # MCP tool implementations
    │   ├── __init__.py
    │   ├── connection.py               # Connection management tools
    │   ├── navigation.py               # Score navigation tools
    │   ├── notes_measures.py           # Note and measure manipulation
    │   ├── sequences.py                # Batch operation tools
    │   ├── staff_instruments.py        # Staff and instrument tools
    │   └── time_tempo.py               # Timing and tempo tools
    └── types/                          # Type definitions
        ├── __init__.py
        └── action_types.py             # WebSocket action type definitions

Requirements

Create a requirements.txt file with:

fastmcp
websockets

MIDI Pitch Reference

Common MIDI pitch values for reference:

  • Middle C: 60
  • C Major Scale: 60, 62, 64, 65, 67, 69, 71, 72
  • Chromatic: C=60, C#=61, D=62, D#=63, E=64, F=65, F#=66, G=67, G#=68, A=69, A#=70, B=71

Duration Reference

Duration format: {"numerator": int, "denominator": int}

  • Whole note: {"numerator": 1, "denominator": 1}
  • Half note: {"numerator": 1, "denominator": 2}
  • Quarter note: {"numerator": 1, "denominator": 4}
  • Eighth note: {"numerator": 1, "denominator": 8}
  • Dotted quarter: {"numerator": 3, "denominator": 8}

推荐服务器

Baidu Map

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。

官方
精选
JavaScript
Playwright MCP Server

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。

官方
精选
TypeScript
Magic Component Platform (MCP)

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。

官方
精选
本地
TypeScript
Audiense Insights MCP Server

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。

官方
精选
本地
TypeScript
VeyraX

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。

官方
精选
本地
graphlit-mcp-server

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。

官方
精选
TypeScript
Kagi MCP Server

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。

官方
精选
Python
e2b-mcp-server

e2b-mcp-server

使用 MCP 通过 e2b 运行代码。

官方
精选
Neon MCP Server

Neon MCP Server

用于与 Neon 管理 API 和数据库交互的 MCP 服务器

官方
精选
Exa MCP Server

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。

官方
精选