InsightChat

InsightChat is a real-time chat application built with the MERN stack (MongoDB, Express, React/Angular, Node.js) and powered by Google Gemini AI. It features real-time messaging, AI-assisted writing, and user authentication.

Project Structure

The project is divided into two main parts:

• chat-server: The backend Node.js/Express server.
• chat-client: The frontend Angular application.

Prerequisites

Before you begin, ensure you have the following installed:

• Node.js (v14 or higher recommended)
• Angular CLI (npm install -g @angular/cli)
• MongoDB (Local installation or Atlas URI)

Installation

1. Server Setup

Navigate to the chat-server directory and install dependencies:

cd chat-server
npm install

2. Client Setup

Navigate to the chat-client directory and install dependencies:

cd chat-client
npm install

Configuration

You need to configure the environment variables for the server.

1. Create a .env file in the chat-server directory.
2. Add the following variables:

MONGO_URL=your_mongodb_connection_string
PORT=3000
GEMINI_API_KEY=your_google_gemini_api_key
JWT_SECRET=your_jwt_secret_key

Variable	Description
MONGO_URL	Connection string for your MongoDB database.
PORT	Port for the server to run on (default: 3000).
GEMINI_API_KEY	API key for Google Gemini (required for AI features).
JWT_SECRET	Secret key used for signing JSON Web Tokens.

Running the Application

Start the Server

In the chat-server directory, run:

# Production start
npm start

# Or directly with node
node index.js

The server will start at http://localhost:3000 (or your configured port). Socket.io will also listen on this port.

Start the Client

In the chat-client directory, run:

ng serve

Navigate to http://localhost:4200/ in your browser. The application will automatically reload if you change any of the source files.

Features

• Real-time Messaging: ID-based socket rooms for private chatting.
• AI Integration:
  • Chat with "AI Bot" for queries.
  • AI-powered message rewriting (Formal, Casual, Grammar fixes).
• Authentication: Secure signup and login with JWT.
• Message Status: Real-time delivered and read receipts.

Troubleshooting

• CORS Issues: If you experience connection issues between client and server, ensure cors is configured correctly in chat-server/index.js to allow http://localhost:4200.
• MongoDB Connection: Ensure your IP is whitelisted if using MongoDB Atlas, or that your local MongoDB service is running.
• Gemini API: If AI features fail, verify your GEMINI_API_KEY is valid and has quota.