Regennexus Uap

What is RegenNexus UAP?

RegenNexus UAP (Universal Adapter Protocol) connects devices, robots, apps, and AI agents with minimal latency and maximum security. It is MCP-compatible for seamless AI integration, targeting developers building IoT, robotics, and AI-driven systems.

How to use RegenNexus UAP?

Install via pip (pip install regennexus[full]), then import RegenNexusProtocol, initialize, register entities, and send messages. The package also provides a CLI (regen server, regen run) and an MCP server for use with Claude Desktop (python -m regennexus.mcp_server).

Key features of RegenNexus UAP

• Supports Raspberry Pi, Arduino, NVIDIA Jetson, and IoT devices
• Integrates robotic arms like Amber B1 and Lucid One
• Multiple transport layers: IPC, UDP Multicast, WebSocket, Message Queue
• AES-128/256-GCM encryption, ECDH-384 key exchange, token/API key auth
• MCP server for AI agent control via Claude Desktop
• LLM Bridge for Ollama and LM Studio integration
• Mesh networking with auto-discovery across devices

Use cases of RegenNexus UAP

• Control robot arms and IoT hardware from an AI assistant
• Build secure, low‑latency communication between distributed devices
• Connect local LLMs (e.g., Ollama, LM Studio) to physical hardware
• Create ad‑hoc mesh networks for sensor arrays or edge nodes

FAQ from RegenNexus UAP

What runtime dependencies does it require?

Python 3.8+ and pip. Optional extras include FastAPI ([api]), MQTT ([mqtt]), robotics ([robotics]), and Arduino ([arduino]).

How does MCP integration work?

Start the MCP server with python -m regennexus.mcp_server, then configure it in claude_desktop_config.json. Claude can then send commands directly to connected hardware.

What transport layers are available and their typical latency?

IPC (<0.1ms), UDP Multicast (1-5ms), WebSocket (10-50ms), and Message Queue (variable). Each is suited for local processes, LAN discovery, remote/internet, or reliable delivery respectively.

Where does data live and is authentication supported?

Data flows between registered entities; the protocol itself does not persist data. Authentication uses tokens, API keys, adaptive rate limiting, and ECDH-384 key exchange for session encryption.

Are there any known limits?

The README does not list specific limits but notes that latency depends on the transport layer chosen and that optional dependencies may be required for certain hardware integrations.