Industrial
FreeNot checkedAn open-source MCP server that bridges AI models with industrial equipment, supporting multiple protocols like Modbus, OPC UA, and MQTT for reading data and con
About
An open-source MCP server that bridges AI models with industrial equipment, supporting multiple protocols like Modbus, OPC UA, and MQTT for reading data and controlling machines.
README
🏭 EDGEAI MCP
The Open-Source Bridge Between AI and Industrial Equipment
Features • Quick Start • Protocols • Docs • Contributing
🌟 What is Industrial MCP?
Industrial MCP is an open-source project that implements the Model Context Protocol (MCP) for industrial environments. It allows any AI model (Claude, GPT, Mistral, Llama, etc.) to read data from and control industrial equipment through a standardized interface.
Think of it as USB-C for Industrial AI — one universal connector for all your machines.
The Problem We Solve
❌ Traditional Approach:
AI Model ←→ Custom API 1 ←→ Machine 1 (Modbus)
AI Model ←→ Custom API 2 ←→ Machine 2 (OPC UA)
AI Model ←→ Custom API 3 ←→ Machine 3 (MQTT)
= N×M integration nightmare 😱
✅ With Industrial MCP:
AI Model ←→ MCP Protocol ←→ Industrial MCP Server ←→ Any Machine
= One standard interface for everything 🎉
✨ Features
| Feature | Description |
|---|---|
| 🔌 Multi-Protocol Support | Modbus TCP/RTU, OPC UA, MQTT, S7 (Siemens) |
| 🤖 AI-Ready | Works with Claude, ChatGPT, Mistral, local LLMs |
| 📍 Edge-First | Runs on Raspberry Pi, Jetson, any Linux device |
| 🔒 Data Sovereignty | All processing on-premise, no cloud required |
| 🇪🇺 GDPR Compliant | Data never leaves your factory |
| 💬 Natural Language | Talk to your machines in plain language |
| 📊 Built-in Monitoring | Real-time dashboards and alerts |
| 🔧 Extensible | Add custom protocols with simple Python plugins |
🚀 Quick Start
Installation
# Clone the repository
git clone https://github.com/YOUR_USERNAME/industrial-mcp.git
cd industrial-mcp
# Create virtual environment
python -m venv venv
source venv/bin/activate # On Windows: venv\Scripts\activate
# Install dependencies
pip install -e .
Basic Usage
from industrial_mcp import MCPServer, ModbusAdapter
# Create MCP server
server = MCPServer(name="my-factory")
# Connect to a Modbus device (e.g., a pump)
pump = ModbusAdapter(
host="192.168.1.100",
port=502,
device_name="pump-01"
)
# Register the device
server.register_device(pump)
# Define tools that AI can use
@server.tool("get_pump_status")
async def get_pump_status():
"""Get the current status of the main pump"""
temp = await pump.read_register(address=100)
vibration = await pump.read_register(address=101)
return {
"temperature": temp,
"vibration": vibration,
"status": "normal" if vibration < 50 else "warning"
}
# Start the MCP server
server.run(host="0.0.0.0", port=8080)
Connect with Claude Desktop
Add to your Claude Desktop config (claude_desktop_config.json):
{
"mcpServers": {
"industrial": {
"command": "python",
"args": ["-m", "industrial_mcp", "--config", "config.yaml"]
}
}
}
Now you can ask Claude:
"What's the temperature of pump-01?"
"Is the vibration level normal?"
"Show me the status of all connected devices."
📡 Supported Protocols
| Protocol | Status | Use Case |
|---|---|---|
| Modbus TCP | ✅ Stable | PLCs, sensors, meters |
| Modbus RTU | ✅ Stable | Serial devices, RS-485 |
| OPC UA | ✅ Stable | Modern industrial systems |
| MQTT | ✅ Stable | IoT sensors, lightweight devices |
| Siemens S7 | 🔄 Beta | Siemens PLCs (S7-300/400/1200/1500) |
| BACnet | 📋 Planned | Building automation |
| EtherNet/IP | 📋 Planned | Allen-Bradley, Rockwell |
🏗️ Architecture
┌─────────────────────────────────────────────────────────────┐
│ YOUR FACTORY │
│ │
│ ┌──────────┐ ┌──────────┐ ┌──────────┐ ┌──────────┐ │
│ │ Machine 1│ │ Machine 2│ │ Machine 3│ │ Sensor N │ │
│ │ (Modbus) │ │ (OPC UA) │ │ (MQTT) │ │ (Modbus) │ │
│ └────┬─────┘ └────┬─────┘ └────┬─────┘ └────┬─────┘ │
│ │ │ │ │ │
│ └─────────────┴──────┬──────┴─────────────┘ │
│ │ │
│ ┌────────▼────────┐ │
│ │ INDUSTRIAL MCP │ │
│ │ SERVER │ ← Runs on Edge │
│ │ ┌────────────┐ │ │
│ │ │ Adapters │ │ │
│ │ │ ┌──────┐ │ │ │
│ │ │ │Modbus│ │ │ │
│ │ │ │OPC UA│ │ │ │
│ │ │ │ MQTT │ │ │ │
│ │ │ └──────┘ │ │ │
│ │ └────────────┘ │ │
│ │ ┌────────────┐ │ │
│ │ │ MCP Server │ │ │
│ │ └────────────┘ │ │
│ └────────┬────────┘ │
│ │ │
└────────────────────────────┼────────────────────────────────┘
│ MCP Protocol (JSON-RPC)
▼
┌─────────────────┐
│ AI CLIENT │
│ Claude / GPT │
│ Local LLM │
└─────────────────┘
💡 Use Cases
🔧 Predictive Maintenance
@server.tool("analyze_vibration_pattern")
async def analyze_vibration():
"""Compare current vibration with historical failure patterns"""
current = await pump.read_vibration()
historical = await db.get_failure_patterns()
similarity = calculate_similarity(current, historical)
return {
"similarity_to_failure": f"{similarity}%",
"recommendation": "Schedule inspection" if similarity > 80 else "Normal"
}
📊 Real-time Monitoring
@server.tool("get_production_status")
async def get_production():
"""Get real-time production line status"""
return {
"units_produced": await plc.read("production_count"),
"efficiency": await calculate_oee(),
"downtime_minutes": await get_downtime()
}
🚨 Anomaly Detection
@server.tool("check_anomalies")
async def check_anomalies():
"""Detect anomalies across all connected devices"""
anomalies = []
for device in server.devices:
if await device.is_anomalous():
anomalies.append(device.name)
return {"anomalies": anomalies, "count": len(anomalies)}
🖥️ Edge Deployment
Raspberry Pi 4/5
# Install on Raspberry Pi
curl -sSL https://get.industrial-mcp.io | bash
# Or manually
pip install industrial-mcp[raspberry]
NVIDIA Jetson
# Optimized for Jetson with local LLM support
pip install industrial-mcp[jetson]
Docker
docker run -d \
--name industrial-mcp \
-p 8080:8080 \
-v ./config.yaml:/app/config.yaml \
industrialmcp/server:latest
📖 Documentation
| Document | Description |
|---|---|
| Getting Started | First steps with Industrial MCP |
| Configuration | YAML configuration reference |
| Adapters Guide | How to use protocol adapters |
| Custom Adapters | Write your own adapter |
| Security | Authentication and encryption |
| API Reference | Complete API documentation |
🤝 Contributing
We welcome contributions! See CONTRIBUTING.md for guidelines.
Development Setup
# Clone and install dev dependencies
git clone https://github.com/YOUR_USERNAME/industrial-mcp.git
cd industrial-mcp
pip install -e ".[dev]"
# Run tests
pytest
# Run linting
ruff check .
Roadmap
- Modbus TCP/RTU adapter
- OPC UA adapter
- MQTT adapter
- Basic MCP server
- Siemens S7 adapter (in progress)
- Web dashboard
- Local LLM integration (Ollama)
- Anomaly detection ML models
- ATEX certification support
🌍 Community
- 💬 Discord
- 📧 Mailing List
📜 License
This project is licensed under the Apache License 2.0 - see the LICENSE file for details.
🏢 Enterprise Edition
Need more features for your enterprise?
| Feature | Open Source | Enterprise |
|---|---|---|
| Core MCP Server | ✅ | ✅ |
| Modbus/OPC UA/MQTT | ✅ | ✅ |
| Community Support | ✅ | ✅ |
| Multi-site Management | ❌ | ✅ |
| Advanced Analytics Dashboard | ❌ | ✅ |
| Priority Support (SLA) | ❌ | ✅ |
| CE/ATEX Certification Kit | ❌ | ✅ |
| Custom Protocol Development | ❌ | ✅ |
📧 Contact: [email protected]
Made with ❤️ in France 🇫🇷
Part of the La French Tech ecosystem
Installing Industrial
This server has no published package — it is built from source. Open the repository and follow its README.
▸ github.com/xiao98/EdgeAI-MCPFAQ
Is Industrial MCP free?
Yes, Industrial MCP is free — one-click install via Unyly at no cost.
Does Industrial need an API key?
No, Industrial runs without API keys or environment variables.
Is Industrial hosted or self-hosted?
Self-hosted: the server runs locally on your machine via the install command above.
How do I install Industrial in Claude Desktop, Claude Code or Cursor?
Open Industrial on unyly.org, pick your client tab (Claude Desktop, Claude Code, Cursor) and press Install — the config is generated automatically, no JSON editing.
Related MCPs
Fetch
Web content fetching and conversion for efficient LLM usage.
AWS KB Retrieval
Retrieval from AWS Knowledge Base using Bedrock Agent Runtime.
by modelcontextprotocolSpring AI MCP Server
Provides auto-configuration for setting up an MCP server in Spring Boot applications.
llm-analysis-assistant
A very streamlined mcp client that supports calling and monitoring stdio/sse/streamableHttp, and can also view request responses through the /logs page. It also
by xuzexin-hzCompare Industrial with
Not sure what to pick?
Find your stack in 60 seconds
Author?
Embed badge for your README
Browse similar
All ai MCPs
