Agent Initializr

Agent Initializr is a project that helps you scaffold scalable AI agent backends in JavaScript/TypeScript using NestJS and LangGraph. It generates a production-ready backend that you can use as-is or extend with your own core business logic.

Below is the structure of the generated project:

AI Agent Backend Template

A beginner-friendly NestJS template for building AI agents with streaming capabilities and conversation management.

Features

• 🤖 Ready-to-use AI agent implementation
• 🌊 Real-time streaming responses
• 💾 Conversation history management
• 🔄 Support for multiple LLM providers (OpenAI, Google)
• 📡 Built-in Redis pub/sub for real-time messaging
• 🎯 Clean and maintainable architecture

Prerequisites to use the generated project

• Node.js (v18 or higher)
• Docker and Docker Compose (for local development)
• OpenAI API key or Google AI API key

Quick Start

1. Start the required services (Redis and PostgreSQL) using Docker Compose:

docker compose up -d

This will start:

• PostgreSQL at localhost:5433
• Redis at localhost:6379

You can check the status of the services with:

docker compose ps

1. Clone the repository
2. Install dependencies:

pnpm install

3. Create a .env file in the root directory:

# Choose your model provider
MODEL_PROVIDER=GOOGLE  # or OPENAI

# For Google AI
GOOGLE_GENAI_API_KEY=your_api_key
GOOGLE_GENAI_MODEL=gemini-pro

# For OpenAI
OPENAI_API_KEY=your_api_key
OPENAI_MODEL=gpt-3.5-turbo

# Redis Configuration
REDIS_HOST=localhost
REDIS_PORT=6379
REDIS_USERNAME=default
REDIS_PASSWORD=

# App Configuration
PORT=3001

4. Start the development server:

pnpm run start:dev

API Endpoints

• POST /api/agent/chat - Send a message to the agent
• GET /api/agent/stream - Stream agent responses (SSE)
• GET /api/agent/history/:threadId - Get conversation history

Basic Usage Example

// Chat endpoint
await fetch('http://localhost:3001/api/agent/chat', {
  method: 'POST',
  headers: { 'Content-Type': 'application/json' },
  body: JSON.stringify({
    threadId: 'unique-thread-id',
    content: [{ type: 'text', text: 'Hello, AI!' }],
    type: 'human'
  })
});

// Stream endpoint (using EventSource)
const sse = new EventSource('http://localhost:3001/api/agent/stream?threadId=unique-thread-id&content=Hello');
sse.onmessage = (event) => {
  const data = JSON.parse(event.data);
  console.log(data.content);
};

Project Structure

src/
├── agent/              # AI agent implementation
├── api/               # HTTP endpoints and DTOs
└── messaging/         # Redis messaging service

Contributing

Contributions are welcome! Just DM me on X/Twitter