RegenNexus UAP
Universal Adapter Protocol - Connect devices, robots, apps, and AI agents with minimal latency and maximum security. MCP-compatible for seamless AI integration.

Installation
pip install regennexus
Or with all features:
pip install regennexus[full]
Quick Start
import asyncio
from regennexus import RegenNexusProtocol
async def main():
protocol = RegenNexusProtocol()
await protocol.initialize()
# Register entities
await protocol.registry.register_entity(
entity_id="sensor_01",
entity_type="device",
capabilities=["temperature", "humidity"]
)
# Send messages
await protocol.send_message(
sender="controller",
recipient="sensor_01",
intent="read",
payload={"sensors": ["temperature"]}
)
await protocol.shutdown()
asyncio.run(main())
Features
Device Support
- Raspberry Pi - GPIO, PWM, camera, sensors
- Arduino - Digital/analog I/O, serial commands
- NVIDIA Jetson - GPU, CUDA, camera, inference
- IoT Devices - MQTT, HTTP, CoAP protocols
Robotic Arms
- Amber B1 - 7-DOF control, gripper, trajectory
- Lucid One - Cartesian control, force sensing, teach mode
from regennexus.plugins import get_amber_b1_plugin
AmberB1 = get_amber_b1_plugin()
arm = AmberB1(entity_id="arm_001", mock_mode=True)
await arm.initialize()
# Move joints
await arm.move_to([0, 45, -30, 0, 90, 0, 0], duration=2.0)
# Gripper control
await arm.open_gripper()
await arm.close_gripper(force=15.0)
Transport Layers
| Transport | Latency | Use Case |
|---|---|---|
| IPC | < 0.1ms | Local processes |
| UDP Multicast | 1-5ms | LAN discovery |
| WebSocket | 10-50ms | Remote/internet |
| Message Queue | Variable | Reliable delivery |
Security
- Encryption: AES-128/256-GCM
- Key Exchange: ECDH-384
- Authentication: Tokens, API keys
- Rate Limiting: Adaptive throttling
AI Integration (MCP)
Control hardware directly from Claude Desktop or any MCP-compatible AI:
# Start MCP server for Claude Desktop
python -m regennexus.mcp_server
Configure in claude_desktop_config.json:
{
"mcpServers": {
"regennexus": {
"command": "python",
"args": ["-m", "regennexus.mcp_server"]
}
}
}
Now ask Claude:
- "Move the robot arm to pick position"
- "Turn on GPIO pin 17"
- "What's the temperature sensor reading?"
LLM Bridge (Ollama, LM Studio)
Connect local LLMs to hardware:
from regennexus.bridges import LLMBridge, LLMConfig
llm = LLMBridge(LLMConfig(provider="ollama", model="llama3"))
response = await llm.chat("Turn on the lights")
Mesh Networking
Auto-discovery across devices on the network:
from regennexus.core import MeshNetwork, MeshConfig
mesh = MeshNetwork(MeshConfig(
node_id="controller",
capabilities=["command"]
))
await mesh.start()
# Devices auto-discovered
for peer in mesh.get_peers():
print(f"Found: {peer.node_id} ({peer.capabilities})")
Interactive Demos
Try RegenNexus in Google Colab:
- Basic Demo - Core communication
- Security Demo - Encryption & auth
CLI Usage
# Start server
regen server --host 0.0.0.0 --port 8080
# Run example
regen run examples/robotic_arms/arm_demo.py
# Version info
regen version
Optional Dependencies
pip install regennexus[api] # FastAPI server
pip install regennexus[mqtt] # MQTT support
pip install regennexus[robotics] # Robotic arms
pip install regennexus[arduino] # Arduino support
pip install regennexus[dev] # Development tools
Documentation
- Getting Started
- MCP Integration
- Device Integration
- Robotic Arms Guide
- Security Guide
- API Reference
- ROS Integration
Examples
examples/
├── simple_connection/ # Basic protocol usage
├── mcp_integration/ # Claude Desktop & LLM demos
├── mesh_network/ # Device auto-discovery
├── llm_integration/ # Ollama/LM Studio demos
├── robotic_arms/ # Amber B1 & Lucid One demos
├── ros_integration/ # ROS 2 bridge examples
├── security/ # Encryption & auth
└── binder/ # Jupyter notebooks
Docker
docker-compose up
See Docker Deployment for details.
Contributing
We welcome contributions! See CONTRIBUTING.md for guidelines.
License
MIT License - see LICENSE for details.
RegenNexus UAP - Connect Everything, Securely.
Copyright (c) 2024-2025 ReGen Designs LLC