Arduino MCP Server (Simple)
Enables AI agents to interact with Arduino boards for compiling, uploading sketches, and serial communication.
README
Arduino MCP Server (Simple)
A simple Model Context Protocol (MCP) server that enables AI agents to interact with Arduino boards for development and testing.
Note: This project was developed and tested on macOS with Cursor IDE. While it should work on other platforms, some paths and configurations may need adjustment for Windows or Linux systems.
What is this?
This project demonstrates how to create an MCP server that allows AI agents to:
- List available Arduino boards
- Compile Arduino sketches
- Upload sketches to boards
- Send/receive serial messages
Prerequisites
- Arduino CLI installed and configured
- Python 3.10+
- An Arduino board connected via USB
Installation
- Install the package directly from GitHub:
# Install from main branch
pip install git+https://github.com/amahpour/arduino-mcp-server-simple.git
# Or install from the specific MCP-enabled branch
pip install git+https://github.com/amahpour/arduino-mcp-server-simple.git@codex/update-pyproject.toml-for-mcp-tool
# For development (editable install)
git clone https://github.com/amahpour/arduino-mcp-server-simple.git
cd arduino-mcp-server-simple
pip install -e .
- Make sure Arduino CLI is installed and configured:
arduino-cli version
Usage
Start the MCP Server
After installation, you can start the server in multiple ways:
# Using the console script (recommended)
arduino-mcp-tool
# Or using module execution
python -m arduino_mcp_tool
For development or local testing:
# From the repository directory
python -m arduino_mcp_tool
Package Structure
The project is organized as a proper Python package:
arduino_mcp_tool/
├── __init__.py # Main MCP server implementation
└── __main__.py # Entry point for python -m execution
Cursor Integration
This project includes a .cursor/mcp.json configuration file that allows Cursor users to integrate the Arduino MCP server directly into their development environment.
The configuration automatically registers the Arduino MCP server, giving you access to Arduino development tools directly within Cursor's AI features. You can:
- Ask Cursor to compile and upload Arduino sketches
- Test serial communication through the AI interface
- Get help with Arduino development workflows
The MCP server will be available as soon as you open this project in Cursor.
Available Tools
list_ports()- List all available USB/serial portsping()- Simple health checkcompile(sketch, fqbn=None, port=None)- Compile an Arduino sketchupload(sketch, port, fqbn=None)- Upload a sketch to a boardserial_send(port, baud, message, timeout=2)- Send a message over serial and read responseserial_write(port, baud, message)- Send a message over serial without reading responseserial_read(port, baud, timeout=2)- Read incoming data from serial without sending anything
Example Workflow
- List available ports to find your Arduino
- Compile your sketch (auto-detects board type)
- Upload to the board
- Send test messages over serial
Example Sketch
The included sketches/echo_serial/echo_serial.ino is a simple echo server that:
- Receives serial input
- Echoes it back with a prefix
- Perfect for testing the MCP server functionality
Integration
This MCP server can be integrated with any MCP-compatible AI agent or development environment to enable automated Arduino development workflows.
MCP Configuration
When using this server with MCP clients, you can configure it as:
{
"mcpServers": {
"arduino": {
"command": "arduino-mcp-tool",
"args": []
}
}
}
Or if installed in development mode:
{
"mcpServers": {
"arduino": {
"command": "python",
"args": ["-m", "arduino_mcp_tool"]
}
}
}
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。