WinCC OA MCP Server
A Model Context Protocol server that exposes WinCC OA datapoint operations as LLM-callable tools, enabling read, write, create, delete, and query operations on datapoints.
README
WinCC OA MCP Server
A Model Context Protocol (MCP) server that exposes WinCC OA datapoint operations as LLM-callable tools. It runs inside a WinCC OA Node.js Manager and uses the native winccoa-manager add-on to interact with the WinCC OA runtime.
Features
| Tool | Description | Hint |
|---|---|---|
winccoa_dp_get |
Read one or more datapoint element values | read-only |
winccoa_dp_set |
Write values (fire-and-forget or confirmed) | write |
winccoa_dp_create |
Create a new datapoint | write |
winccoa_dp_delete |
Delete a datapoint | destructive |
winccoa_dp_copy |
Copy a datapoint to a new name | write |
winccoa_dp_names |
List DPs matching a pattern | read-only |
winccoa_dp_exists |
Check whether a DP identifier exists | read-only |
winccoa_dp_query |
Run a WinCC OA SQL-like query | read-only |
winccoa_dp_types |
List datapoint types matching a pattern | read-only |
Prerequisites
- WinCC OA 3.20 (or compatible) with the Node.js Manager enabled
- Node.js 18+ (bundled with WinCC OA or system-installed)
- The
winccoa-managernative add-on (provided by the WinCC OA runtime – not on npm)
Installation
cd winccoa-mcp-server
npm install
npm run build
This compiles TypeScript sources from src/ into dist/.
Configuration
Since the WinCC OA Node.js Manager does not support command-line arguments, all configuration is done through a .env file placed next to the built entry point (or in the working directory).
A default .env file is included in the repository:
# Transport mode: "stdio" or "http"
MCP_TRANSPORT=stdio
# Port for the HTTP transport (only used when MCP_TRANSPORT=http)
MCP_HTTP_PORT=47899
# Maximum response size in characters to prevent overwhelming the LLM context
MCP_CHARACTER_LIMIT=25000
| Variable | Description | Default |
|---|---|---|
MCP_TRANSPORT |
"stdio" or "http" |
stdio |
MCP_HTTP_PORT |
Port for HTTP transport | 47899 |
MCP_CHARACTER_LIMIT |
Max response length (chars) | 25000 |
Usage
stdio transport (default)
The server communicates over stdin / stdout. Use this when the MCP client (e.g. Claude Desktop, VS Code) starts the process directly or when testing with MCP Inspector.
node dist/index.js
HTTP transport
Set MCP_TRANSPORT=http in your .env file, then start the server. It will listen on the port specified by MCP_HTTP_PORT.
node dist/index.js
Integration with WinCC OA
This server is designed to be run as customer code inside a WinCC OA Node.js Manager. The typical setup:
- Place the built
dist/folder (or the whole project) in your WinCC OA project's scripts directory. - Configure a Node.js Manager in the WinCC OA console with this server as the entry script.
- The WinCC OA bootstrap will load the native
winccoa-manageradd-on, then executedist/index.js.
MCP client configuration
Claude Desktop (claude_desktop_config.json)
{
"mcpServers": {
"winccoa": {
"command": "node",
"args": ["C:/path/to/winccoa-mcp-server/dist/index.js"]
}
}
}
VS Code (.vscode/mcp.json)
{
"servers": {
"winccoa": {
"type": "stdio",
"command": "node",
"args": ["C:/path/to/winccoa-mcp-server/dist/index.js"]
}
}
}
For HTTP transport, point the client to http://localhost:47899/mcp (or your configured port).
Development
# Run in development mode (tsx, no build step)
npm run dev
# Test with MCP Inspector
npm run inspect
To switch transports during development, edit the .env file.
Project structure
src/
├── index.ts # Entry point – transport setup
├── server.ts # McpServer creation
├── constants.ts # Shared constants
├── winccoa-client.ts # WinccoaManager singleton
├── types/
│ └── winccoa-manager.d.ts # Type declarations for native add-on
├── tools/
│ ├── register-all.ts # Tool registration orchestrator
│ ├── dp-get.ts # winccoa_dp_get
│ ├── dp-set.ts # winccoa_dp_set
│ ├── dp-create.ts # winccoa_dp_create
│ ├── dp-delete.ts # winccoa_dp_delete
│ ├── dp-copy.ts # winccoa_dp_copy
│ ├── dp-names.ts # winccoa_dp_names
│ ├── dp-exists.ts # winccoa_dp_exists
│ ├── dp-query.ts # winccoa_dp_query
│ └── dp-types.ts # winccoa_dp_types
└── utils/
├── error-handler.ts # WinCC OA error handling
└── formatters.ts # Response formatting helpers
Adding new tools
- Create
src/tools/your-tool.tsfollowing the existing pattern (Zod schema +registerTool()). - Import and call your registration function in
src/tools/register-all.ts. - Rebuild with
npm run build.
License
MIT
Recommended Servers
playwright-mcp
A Model Context Protocol server that enables LLMs to interact with web pages through structured accessibility snapshots without requiring vision models or screenshots.
Magic Component Platform (MCP)
An AI-powered tool that generates modern UI components from natural language descriptions, integrating with popular IDEs to streamline UI development workflow.
Audiense Insights MCP Server
Enables interaction with Audiense Insights accounts via the Model Context Protocol, facilitating the extraction and analysis of marketing insights and audience data including demographics, behavior, and influencer engagement.
VeyraX MCP
Single MCP tool to connect all your favorite tools: Gmail, Calendar and 40 more.
graphlit-mcp-server
The Model Context Protocol (MCP) Server enables integration between MCP clients and the Graphlit service. Ingest anything from Slack to Gmail to podcast feeds, in addition to web crawling, into a Graphlit project - and then retrieve relevant contents from the MCP client.
Kagi MCP Server
An MCP server that integrates Kagi search capabilities with Claude AI, enabling Claude to perform real-time web searches when answering questions that require up-to-date information.
E2B
Using MCP to run code via e2b.
Neon Database
MCP server for interacting with Neon Management API and databases
Exa Search
A Model Context Protocol (MCP) server lets AI assistants like Claude use the Exa AI Search API for web searches. This setup allows AI models to get real-time web information in a safe and controlled way.
Qdrant Server
This repository is an example of how to create a MCP server for Qdrant, a vector search engine.