Chat Lounge

A full-featured AI chat application built with SvelteKit, supporting multiple AI providers, conversation management, and advanced features like TTS and voice recording.

Features

🤖 AI Integration

• Multiple AI Providers: OpenAI, Anthropic, Mistral, Google
• Streaming Responses: Real-time chat with proper error handling
• Custom Agents: Create AI personalities with custom instructions
• Per-user API Keys: Secure API key management

💬 Conversation Management

• Branching/Threading: Parent-child message relationships
• Message Editing: Edit and regenerate messages
• Organization: Pin conversations and organize with folders
• Sharing: Public conversation sharing
• Search: Full-text search across conversations and messages

🔊 Audio Features

• Text-to-Speech: ElevenLabs integration with voice selection
• Voice Recording: Built-in voice recording capabilities
• Audio Controls: Playback controls with audio stores

🏗️ Architecture

• Frontend: SvelteKit with TypeScript
• UI: Shadcn-svelte components with TailwindCSS
• Backend: SvelteKit with adapter-node
• Database: MySQL/MariaDB with Prisma ORM
• Queue: Redis/Valkey with BullMQ for async processing
• Auth: Clerk authentication
• Real-time: Socket.io for live updates

Quick Start

Prerequisites

• Node.js 18+
• pnpm
• MySQL/MariaDB
• Redis/Valkey

Installation

1. Clone the repository:

git clone https://github.com/lordluceus/chat-lounge.git
cd chat-lounge

2. Install dependencies:

pnpm install

3. Set up environment variables:

cp .env.example .env
# Edit .env with your configuration

4. Run database migrations:

pnpm migrate-dev

5. Start development server:

pnpm dev

Development Commands

Core Development

pnpm dev          # Start development server
pnpm build        # Build for production
pnpm preview      # Preview production build

Code Quality

pnpm lint         # Run Prettier and ESLint checks
pnpm format       # Auto-format code
pnpm check        # Run SvelteKit sync and TypeScript checks
pnpm check:watch  # Run checks in watch mode

Database

pnpm migrate-dev     # Run migrations in development
pnpm migrate-prod    # Run migrations in production
pnpm migrate-create  # Create new migration
pnpm studio         # Open Prisma Studio
pnpm prisma-generate # Generate Prisma client

Docker Deployment

The project includes Docker Compose setup with MariaDB and Valkey services:

docker-compose up -d

For production deployment, use the included run.sh script which handles migrations and server startup.

Project Structure

src/
├── lib/
│   ├── components/        # Svelte components
│   ├── server/           # Server-side services
│   │   ├── agents-service.ts
│   │   ├── conversations-service.ts
│   │   ├── ai-service.ts
│   │   └── ...
│   ├── stores/           # Svelte stores
│   ├── types/            # TypeScript definitions
│   └── utils.ts
├── routes/
│   ├── (app)/           # Main application routes
│   ├── api/             # API endpoints
│   └── auth/            # Authentication pages
└── ...

Key Services

• AI Service: Handles multiple AI provider integration
• Conversations Service: Manages conversations and messages
• Agents Service: CRUD operations for AI agents
• Queue Service: Background job processing with BullMQ
• TTS Service: Text-to-speech functionality

Environment Variables

Required environment variables (see .env.example):

• DATABASE_URL: MySQL/MariaDB connection string
• REDIS_URL: Redis/Valkey connection string
• CLERK_SECRET_KEY: Clerk authentication key
• API keys for AI providers (stored per-user in database)

Contributing

1. Follow the existing code style and conventions
2. Run pnpm lint and pnpm check before committing
3. Write meaningful commit messages
4. Test your changes thoroughly

Tech Stack

• Framework: SvelteKit
• Language: TypeScript
• Styling: TailwindCSS + PostCSS
• UI Components: Shadcn-svelte
• Database: Prisma ORM with MySQL/MariaDB
• Queue: BullMQ with Redis/Valkey
• Authentication: Clerk
• Real-time: Socket.io
• AI SDKs: Vercel AI SDK with multiple providers
• Audio: ElevenLabs for TTS