LocalDude

LocalDude

A Node.js/TypeScript project using the Model Context Protocol (MCP) SDK.

GitHub Repository

This project is hosted on GitHub. To clone the repository:

git clone https://github.com/username/localdude.git
cd localdude

Contributing

We welcome contributions! Please see CONTRIBUTING.md for details on how to contribute to this project.

Installation

npm install

Available Scripts

Development

• npm run dev - Starts the application in development mode with auto-rebuild on file changes
• npm run dev:server - Starts the server in development mode with auto-rebuild on file changes

Build and Run

• npm run build - Compiles TypeScript files to JavaScript
• npm run watch - Compiles TypeScript files in watch mode (without restarting)
• npm run start - Builds and starts the main application
• npm run server - Builds and starts the server

Auto-Rebuild Functionality

This project is configured to automatically rebuild and restart when TypeScript files are changed. This is implemented using:

• nodemon: Monitors file changes and triggers rebuilds
• ts-node: Executes TypeScript files directly

The auto-rebuild behavior is configured in nodemon.json, which watches for changes in:

• All .ts files in the root directory
• All .ts files in the src directory (if it exists)

When changes are detected, the application is automatically rebuilt and restarted.

InfluxDB Integration

This project includes integration with InfluxDB for time-series data storage and querying.

Configuration

To use the InfluxDB integration, you need to set up the following environment variables in your .env file:

INFLUXDB_URL=http://localhost:8086
INFLUXDB_TOKEN=your_influxdb_token
INFLUXDB_ORG=your_organization
INFLUXDB_BUCKET=your_bucket

You can copy these from the .env.example file and replace the placeholder values with your actual InfluxDB connection details.

Available InfluxDB Tools

The server provides the following tools for interacting with InfluxDB:

1. influxdb-query: Execute a custom Flux query

   {
     "query": "from(bucket: \"your_bucket\") |> range(start: -1h) |> filter(fn: (r) => r._measurement == \"your_measurement\")"
   }
   

2. influxdb-latest: Get the latest data point for a measurement and field

   {
     "measurement": "your_measurement",
     "field": "your_field",
     "timeRange": "-1h"  // optional, defaults to "-1h"
   }
   

3. influxdb-range: Get data within a time range

   {
     "measurement": "your_measurement",
     "field": "your_field",
     "start": "-1h",
     "stop": "now()"  // optional, defaults to "now()"
   }
   

These tools can be used through the MCP server interface to query your InfluxDB database.