CTS

You operate the CTS MCP Server to automate Close-to-Shore (CTS) task planning and visualize complex signal structures directly in your development environment. It offers automated task creation in Shrimp MCP and interactive D3 visualizations for signal maps, hop dashboards, and dependency graphs, helping you manage large projects with real-time progress and CTS compliance insights.

How to use

You interact with the CTS MCP Server through an MCP client. Start by configuring the local or remote MCP endpoints, then invoke the available tools to export hop plans to Shrimp, render interactive visualizations, or scan a Godot project for signals. The server communicates over the Model Context Protocol via stdio transport, or through a registered local client that invokes tools by name. Your workflow typically includes exporting a hop plan to Shrimp, rendering artifacts for visualization, and optionally scanning a Godot project to feed up-to-date signal data into the visuals.

For production-ready visualizations, enable the experimental UI if you want the updated dashboards and maps, then start the server. Use the standard start flow in your environment and verify that you can connect with an MCP client or IDE integration.

How to install

Prerequisites: ensure you have Node.js 20+ installed on your system, alongside npm. The server is built with TypeScript and runs via Node.js.

1. Open a terminal and navigate to the CTS MCP project directory.

2. Install dependencies.

c d t s _ m c p
npm install

3. Build the server for production use.

npm run build

4. Start the server in a mode that exposes the MCP interface on stdio. You can enable the experimental UI if you want the enhanced visuals.

export CTS_EXPERIMENTAL_MCP_UI=true
npm start

Configuration and usage tips

Two explicit MCP server configurations are commonly used to connect editors and tools to the CTS MCP Server. The first is a local stdio setup, started from your development environment, and the second is a similar setup for Claude Desktop integration.

{
  "mcpServers": {
    "cts": {
      "type": "stdio",
      "name": "cts",
      "command": "node",
      "args": ["/path/to/cts_mcp/build/index.js"],
      "env": {
        "CTS_EXPERIMENTAL_MCP_UI": "true"
      }
    }
  }
}

Examples of common workflows

Export a hop plan to Shrimp and create tasks, then render a signal map or hop dashboard to verify progress and CTS compliance.

# Start an MCP client session or connect to the CTS MCP server, then call the export tool
# Example command structure is provided by your MCP client integration

Troubleshooting and notes

If you encounter issues with rendering or data, ensure the experimental UI flag is correctly set in your environment when starting the server. For render performance, you can rely on lazy loading and incremental rendering in dashboards.

Appendix: Tools and capabilities

The CTS MCP Server provides tools to export plans to Shrimp, render artifacts like signal maps and hop dashboards, and scan Godot projects for signals. Use the tooling to automate task creation, visualize communications between EventBus and SignalBus, and monitor CTS compliance through the dashboard.

Security and deployment notes

Operate the server in a secured environment and keep your runtime dependencies up to date. The server uses a stdio transport with validation to ensure safe inter-process communication.

Available tools

CTS_Export_to_Shrimp

Converts a hop plan JSON into Shrimp MCP tasks and returns created task IDs and counts.

CTS_Render_Artifact

Renders interactive visual artifacts such as signal maps and hop dashboards for CTS projects.

CTS_Scan_Project_Signals

Scans a Godot project to detect EventBus and SignalBus signal definitions and optionally renders a signal map.