Pythagraph RED

You can run and query the Pythagraph RED MCP Server to fetch graph data, analyze graphs, and present the results as readable tables and summaries. This MCP server connects to the Pythagraph RED API, processes graph information, and exposes tools that help you quickly inspect node/edge distributions, statistics, and metadata without writing custom code.

How to use

Start the server using one of the provided MCP configurations, then you can query it through your MCP client to retrieve detailed graph data or concise graph summaries. You will receive formatted tables and concise insights that you can use for reporting, dashboards, or further analysis.

Primary capabilities you will use are two tools: get_graph_data for in-depth graph information (nodes, edges, statistics, and metadata) and get_graph_summary for a quick overview of counts, distributions, and density. Both tools return structured data that your client can render as tables and charts.

To run the MCP server, choose one of the supported runtime methods shown below and then issue queries to the server from your MCP client using the corresponding tool calls. The server handles API communication, formatting, and error handling so you can focus on your analysis.

{
  "mcpServers": {
    "pythagraph-red": {
      "type": "stdio",
      "command": "npx",
      "args": ["-y", "@modelcontextprotocol/server-pythagraph-red"]
    }
  }
}

How to install

Prerequisites you need to have before you run the MCP server: Node.js with npm and, optionally, Docker if you prefer containerized execution.

Option A: Run via NPX (no local installation required)

{
  "mcpServers": {
    "pythagraph-red": {
      "command": "npx",
      "args": [
        "-y",
        "@modelcontextprotocol/server-pythagraph-red"
      ]
    }
  }
}

Option B: Run via Docker (containerized execution)

{
  "mcpServers": {
    "pythagraph-red": {
      "command": "docker",
      "args": [
        "run",
        "-i",
        "--rm",
        "mcp/pythagraph-red"
      ]
    }
  }
}

Configuration notes

The MCP server provides two main usage vectors ready for you to integrate with your MCP client: fetching detailed graph data and obtaining concise graph summaries. The server formats output into tables and descriptive sections for easy visualization. Ensure your client is prepared to render tables and handle the data shapes described by the tools.

API integration and response format

The server communicates with the Pythagraph RED API to retrieve graph information and returns structured results that include graph identifiers, unit information, and the data rows for tables. Typical responses include details about graphData, node/edge distributions, and metadata.

Error handling and reliability

The server includes robust error handling for common issues such as invalid graph IDs, network timeouts, API response errors, invalid JSON, and connection failures. Timeouts are managed to prevent hanging requests and ensure timely feedback to your MCP client.

Development and testing

Development steps you may follow if you want to build or test locally include installing dependencies, building the project, and running tests. Typical workflows involve npm install, npm run build, and npm test, followed by a local run of the MCP server.

npm install
npm run build
npm test

Notes on output and outputs

Automatic table formatting and statistics calculation are performed to present an at-a-glance view of the graph data. Detailed views show the first 10 nodes and the first 10 edges to keep the output compact while still providing meaningful examples of the data.

License

This MCP server is released under the MIT License. You are free to use, modify, and distribute the software in accordance with the license terms.

Available tools

get_graph_data

Retrieve detailed graph data from Pythagraph RED API. Returns nodes, edges, statistics, and metadata formatted as tables and descriptions.

get_graph_summary

Get a concise graph overview including counts, distributions, and density. Optional detailed views can be requested.