agora-ai/DEVELOPMENT.md

Development Guide

Prerequisites

• Node.js 18+ installed
• npm or yarn package manager

Project Structure

agora-ai/
├── frontend/          # Vue.js 3 + Vite application
├── backend/           # Node.js + Express API server
├── LICENSE            # MIT License
└── README.md          # Project documentation

Getting Started

1. Backend Setup

cd backend

# Install dependencies
npm install

# Create .env file (copy from .env.example)
cp .env.example .env

# Edit .env and add your Mistral API key
# MISTRAL_API_KEY=your_key_here

# Start the backend server
npm run dev

The backend will run on http://localhost:3000

2. Frontend Setup

cd frontend

# Install dependencies
npm install

# Create .env file (copy from .env.example)
cp .env.example .env

# Start the development server
npm run dev

The frontend will run on http://localhost:5173

Available Scripts

Backend

• npm start - Start the server
• npm run dev - Start the server with auto-reload (Node.js --watch)

Frontend

• npm run dev - Start development server with HMR
• npm run build - Build for production
• npm run preview - Preview production build

API Endpoints

Health Check

GET /api/health

Create Debate

POST /api/debate
Body: { "prompt": "Your project description" }

Get Debate

GET /api/debate/:id

Add Response

POST /api/debate/:id/response
Body: { "agentRole": "architect", "content": {...} }

WebSocket Connection

The backend supports WebSocket connections for real-time updates. Connect to:

ws://localhost:3000

Database

The project uses SQLite for data storage. The database file is automatically created at:

backend/data/agora.db

Schema:

• debates - Stores project prompts and debate metadata
• responses - Stores AI agent responses for each debate

Technology Stack

Frontend

• Vue.js 3 (Composition API)
• Vite
• Pinia (State Management)
• Radix Vue (Headless UI Components)
• Mermaid.js (Diagrams)
• VueUse (Utilities)

Backend

• Node.js + Express
• WebSocket (ws)
• SQLite (better-sqlite3)
• Helmet (Security)
• CORS
• Express Rate Limit

Next Steps

1. Implement Mistral AI integration for agent responses
2. Add WebSocket real-time communication
3. Implement Mermaid diagram rendering
4. Add voting and consensus mechanisms
5. Create export functionality (JSON, Markdown, PDF)