eigen-plc-mcp

Open-source industrial PLC code generation MCP Server — the open-source version of Siemens Eigen Engineering Agent

Based on the MCP (Model Context Protocol) protocol, it provides intelligent PLC code generation, debugging, and verification capabilities for industrial automation scenarios.

Why this project?

In July 2026, Siemens released Eigen Engineering Agent at WAIC (World Artificial Intelligence Conference) — the world's first industrial AI agent capable of independently writing PLC code. But it:

• ❌ Only supports Siemens' own PLCs (S7-1200/S7-1500)

• ❌ Closed-source, expensive, and not customizable

• ❌ Cannot integrate with third-party AI agent toolchains

What eigen-plc-mcp aims to do:

• ✅ Open-source, free, and customizable

• ✅ Based on the standard MCP protocol, integrable with any AI agent

• ✅ Starting with S7-1200 (TIA Portal) support, expanding step by step

• ✅ Distilling PLC programming experience into a reusable knowledge base

Related MCP server: OpenTIA-MCP

Core Capabilities

Quick Start

Installation

pip install eigen-plc-mcp

Configure MCP Client

Add the following to the MCP configuration in tools such as Claude Code / Cursor / WorkBuddy:

{
  "mcpServers": {
    "eigen-plc-mcp": {
      "command": "python",
      "args": ["-m", "eigen_plc_mcp"],
      "env": {
        "PLC_BRAND": "siemens",
        "PLC_MODEL": "S7-1200"
      }
    }
  }
}

Usage Example

In an AI agent conversation:

请为菠菜初加工产线编写 S7-1200 的清洗段控制逻辑：
- 启动条件：上游传送带运行信号 + 水泵就绪
- 停止条件：急停按钮 + 传送带停止
- 输出：水泵启停 + 传送带速度控制

The MCP Server will automatically generate SCL code that conforms to TIA Portal specifications.

Technical Architecture

┌─────────────────────────────────┐
│         AI Agent (LLM)          │
│    Claude / GPT / Qwen / ...   │
└────────────┬────────────────────┘
             │ MCP Protocol
┌────────────▼────────────────────┐
│      eigen-plc-mcp Server       │
│  ┌──────────┐  ┌──────────────┐ │
│  │Code Gen  │  │ Knowledge    │ │
│  │Engine    │  │ Base         │ │
│  └──────────┘  └──────────────┘ │
│  ┌──────────┐  ┌──────────────┐ │
│  │Validator │  │ Template     │ │
│  │          │  │ Library      │ │
│  └──────────┘  └──────────────┘ │
└────────────┬────────────────────┘
             │
┌────────────▼────────────────────┐
│    TIA Portal / PLC Simulator   │
└─────────────────────────────────┘

Project Background

This project is inspired by the following:

• Siemens Eigen Engineering Agent (China debut at WAIC 2026)

• S7-1200 PLC spinach primary processing line control system (graduation thesis topic)

• MCP protocol ecosystem (the standard bridge between AI agents and industrial systems)

• Pain points in manufacturing AI adoption (high barrier to PLC programming, hard to reuse experience)

Roadmap

Phase 1 — MVP (estimated 30 hours)

• Set up the MCP Server foundation framework

• S7-1200 SCL code template library (washing, conveying, sorting sections)

• Spinach primary processing line knowledge base

• Code generation tool (process description → SCL code)

• Code validation tool (syntax + basic logic checks)

• README + documentation

Phase 2 — Enhancement

• TIA Portal Python API integration

• More production line process templates (food, beverage, packaging)

• PLC code simulation and verification

• Multi-brand PLC adaptation layer

Phase 3 — Productization

• Web UI (process configuration → code generation → download)

• SaaS deployment option

• Enterprise edition (private deployment + customized knowledge base)

Contribution Guidelines

Contributions are welcome! Especially:

• 🏭 PLC programming experience in industrial automation

• 🔧 Development and optimization of the MCP protocol

• 📝 Additions to the production line process knowledge base

• 🧪 Test cases and edge scenarios

License

MIT License

Acknowledgements

• Siemens Eigen Engineering Agent — inspired this open-source alternative

• MCP protocol — Anthropic's open-source AI toolchain standard

• S7-1200 community — countless shared PLC programming experiences

If AI adoption in manufacturing is the future, then open source is the road to that future.