About regennexus-mcp
regennexus-mcp adapts RegenNexus UAP hardware capabilities for AI clients that need to interact with physical devices. It supports local or HTTP-based remote connections, with automatic discovery of installed RegenNexus and mesh nodes.
Key features
- 01GPIO, PWM, and sensor control
- 02Robotic arm and gripper operation
- 03Serial and I2C communication
- 04Camera capture and device details
- 05Mesh node discovery and messaging
- 06Local, remote, and auto modes
Use cases
- 01Read sensors and set GPIO pins
- 02Move a robotic arm or operate a gripper
- 03Capture an image from a camera
- 04Discover mesh nodes and send commands
MCP support
The MCP server exposes RegenNexus hardware and mesh-network functions as tools, including GPIO, robotics, sensors, cameras, and node communication, for MCP clients.
Configuration
Local server config is the same for every client. Copy it into your client’s MCP config file.
uvx regennexus-mcpREGENNEXUS_MODEConnection mode: auto, local, or remoteLocal servers need a runtime on your machine (e.g. Node.js 18+). Config file locations differ by client; see your client’s docs.
Tools
Local servers expose tools only once running on your machine, and the maintainer has not provided a tool list.
Details
From the project READMERegenNexus MCP Server
MCP (Model Context Protocol) server adapter for RegenNexus UAP - enabling AI-controlled hardware.
This package exposes RegenNexus hardware capabilities as MCP tools, allowing Claude Code, Claude Desktop, and other MCP-compatible AI clients to control physical devices.
Features
- Hardware Control: GPIO, PWM, robotic arms, sensors, cameras
- Mesh Networking: Discover and communicate with nodes over LAN (UDP/WebSocket)
- Serial/I2C: Communicate with microcontrollers and sensors
- Two Connection Modes: Local (direct import) or Remote (HTTP API)
- Auto-Discovery: Automatically detects installed RegenNexus and mesh nodes
- MCP Compatible: Works with Claude, Codex, Gemini, and any MCP client
Installation
# MCP server only (for remote UAP connection)
pip install regennexus-mcp
# With local UAP support
pip install regennexus-mcp[local]
# With remote API support
pip install regennexus-mcp[remote]
# Everything
pip install regennexus-mcp[all]
Quick Start
1. Configure Claude Code
Add to your Claude Code MCP settings:
{
"mcpServers": {
"regennexus": {
"command": "regennexus-mcp"
}
}
}
2. Use with Claude
Once configured, Claude can control hardware:
User: "List all connected devices"
Claude: [calls list_devices tool]
User: "Turn on GPIO pin 18"
Claude: [calls gpio_write with pin=18, value=1]
User: "Move the robot arm to position [0, 45, -30, 0, 60, 0, 0]"
Claude: [calls robot_arm_move with positions]
Configuration
Configure via environment variables:
# Connection mode: auto, local, or remote
REGENNEXUS_MODE=auto
# Remote mode settings
REGENNEXUS_ENDPOINT=https://your-uap-server.com
REGENNEXUS_API_KEY=your-api-key
# Local mode settings
REGENNEXUS_CONFIG=/path/to/regennexus-config.yaml
# Logging
REGENNEXUS_LOG_LEVEL=INFO
Claude Code Config with Environment
{
"mcpServers": {
"regennexus": {
"command": "regennexus-mcp",
"env": {
"REGENNEXUS_MODE": "local",
"REGENNEXUS_LOG_LEVEL": "DEBUG"
}
}
}
}
Available Tools (Free Tier)
These tools are provided by the regennexus package (open source):
GPIO & Basic I/O
| Tool | Description |
|---|---|
gpio_write | Set a GPIO pin to HIGH (1) or LOW (0) |
gpio_read | Read the current state of a GPIO pin |
pwm_write | Set PWM duty cycle (0-100%) for motors, LEDs, servos |
Sensors & I2C
| Tool | Description |
|---|---|
read_sensor | Read value from a sensor (temperature, humidity, etc.) |
i2c_scan | Scan I2C bus for connected devices |
Serial Communication
| Tool | Description |
|---|---|
serial_send | Send data over serial port (UART) |
serial_read | Read data from serial port |
Robotics
| Tool | Description |
|---|---|
robot_arm_move | Move a robotic arm to specified joint positions |
gripper_control | Open or close a robotic gripper |
Device Management
| Tool | Description |
|---|---|
list_devices | List all connected hardware devices |
device_info | Get device details (CPU, memory, IP, temperature) |
Camera
| Tool | Description |
|---|---|
camera_capture | Capture a single image from a camera |
Mesh Network
| Tool | Description |
|---|---|
list_nodes | List all nodes in the mesh network |
ping_node | Ping a node and measure network latency |
send_to_node | Send a message/command to a specific node |
broadcast_message | Broadcast a message to all nodes |
find_by_capability | Find nodes with a specific capability |
Premium tools with additional capabilities are available separately.
Connection Modes
Local Mode (Recommended)
Requires regennexus package installed on the same machine. Directly imports UAP modules for minimal latency. No API server needed.
pip install regennexus-mcp[local]
REGENNEXUS_MODE=local regennexus-mcp
This is the recommended mode for most users.
Remote Mode (Advanced)
Connects to a UAP instance running as a service via HTTP API. Requires UAP API server to be running (regen server).
# On the UAP server machine:
regen server --port 8080
# On the Claude Code machine:
pip install regennexus-mcp[remote]
REGENNEXUS_MODE=remote \
REGENNEXUS_ENDPOINT=http://uap-server:8080 \
regennexus-mcp
Auto Mode (Default)
Tries local first, falls back to remote if UAP not installed.
regennexus-mcp # Auto-detects best mode
Running Standalone
# Run MCP server
regennexus-mcp
# Or via Python
python -m regennexus_mcp
Architecture
Claude Code (MCP Client)
│
▼
regennexus-mcp (This Package)
│
├── Local Mode: Direct import
│ │
│ ▼
│ regennexus (UAP)
│
└── Remote Mode: HTTP API
│
▼
UAP Service
Development
# Clone
git clone https://github.com/regennow/regennexus-mcp
cd regennexus-mcp
# Install dev dependencies
pip install -e ".[dev]"
# Run tests
pytest
# Format code
black src/
ruff check src/
Related Projects
- RegenNexus UAP - Universal Adapter Protocol
- MCP Specification - Model Context Protocol
License
MIT License
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.