AI News Article Generator

A sophisticated AI-powered tool that generates professional news articles through a multi-stage process including research, drafting, editing, and finalization.

Features

• Multi-stage Article Generation:

  • Research: AI-powered web browsing for real-time information gathering
  • Drafting: Initial article creation based on research
  • Editing: Content refinement and structure improvement
  • Finalizing: SEO optimization and final polish

• Advanced Web Browsing & Research:

  • Real-time web search across multiple search engines
  • Content extraction from web pages with full text analysis
  • Fact verification using multiple sources
  • Expert source identification and credibility scoring
  • AI-powered insight extraction and analysis
  • Comprehensive research synthesis
  • Configurable time range filtering (day/week/month/year/all)
  • Adjustable source limits (1-15 sources)
  • Source verification and reliability assessment

• Multiple AI Models: Support for various AI models including:

  • OpenAI GPT-4o Mini & GPT-4.1 Mini
  • Anthropic Claude Sonnet 4 & Claude 3.7 Sonnet
  • Google Gemini 2.5 Flash & Gemini 2.5 Pro
  • X.AI Grok 4 Fast

• Customizable Output:

  • News styles (investigative, feature, breaking news, opinion, interview, profile)
  • Tones (formal, casual, humorous, authoritative, conversational, academic)
  • Lengths (short, medium, long)

• SEO Optimization:

  • Keyword integration
  • Meta information generation
  • Source citation

• Real-time Progress: Live updates during article generation

• Source Management: Automatic source gathering and citation

• MongoDB Integration: Article storage and retrieval

Installation

1. Clone the repository:

   git clone <repository-url>
   cd ai-news-article-generator

2. Install dependencies:

   bun install

3. Set up environment variables:

   cp .env.example .env

   Edit .env and add your configuration:

   OPENROUTER_API_KEY=your_openrouter_api_key_here
   MONGODB_URI=mongodb://localhost:27017/article-generator
   PORT=3000
   NODE_ENV=development

4. Start MongoDB (if running locally):

   mongod

Usage

Development Mode

bun run dev

This starts the server with hot reloading on http://localhost:3000.

Production Mode

bun run build
bun run start

Web Interface

Open http://localhost:3000 in your browser to access the web interface where you can:

1. Enter your article topic and summary
2. Select style, tone, and length preferences
3. Choose your preferred AI model
4. Add SEO keywords (optional)
5. Generate the article with real-time progress updates

API Endpoints

Generate Article (Server-Sent Events)

Generates a complete article with real-time progress updates via Server-Sent Events (SSE).

POST /api/articles/generate
Content-Type: application/json

{
  "topic": "Climate Change Impact on Agriculture",
  "summary": "Exploring how climate change affects farming practices worldwide",
  "style": "feature",           // Optional: investigative, feature, breaking-news, opinion, interview, profile
  "tone": "formal",             // Optional: formal, casual, humorous, authoritative, conversational, academic
  "length": "medium",           // Optional: short, medium, long
  "model": "openai/gpt-4-turbo-preview", // Optional: any available model
  "keywords": ["climate change", "agriculture", "farming"] // Optional: array of SEO keywords
}

Response: Server-Sent Events stream with progress updates and final article data.

Enhanced Research Only

Performs comprehensive web-based research without generating an article.

POST /api/articles/research
Content-Type: application/json

{
  "topic": "Renewable Energy Trends 2024",
  "summary": "Current developments in renewable energy sector", // Optional
  "model": "openai/gpt-4-turbo-preview",                     // Optional
  "maxSources": 8,                                           // Optional: 1-15, default 8
  "includeFullContent": true,                                // Optional: default true
  "timeRange": "month"                                       // Optional: day, week, month, year, all
}

Response: JSON object with research results, sources, and AI-generated insights.

Get Available Models

Lists all available AI models and the default model.

GET /api/articles/models

Response:

{
  "models": ["openai/gpt-4-turbo-preview", "anthropic/claude-3-opus", "..."],
  "default": "openai/gpt-4-turbo-preview"
}

Get Article by ID

Retrieves a specific article by its MongoDB ObjectId.

GET /api/articles/:id

Parameters:

• id: MongoDB ObjectId (24-character hexadecimal string)

List Articles (Paginated)

Retrieves a paginated list of articles.

GET /api/articles?limit=10&offset=0

Query Parameters:

• limit: Number of articles to return (default: 10, max: 50)
• offset: Number of articles to skip (default: 0)

Response:

{
  "articles": [...],
  "pagination": {
    "limit": 10,
    "offset": 0,
    "hasMore": true
  }
}

Health Check

Returns service health status and metrics.

GET /api/articles/health

Response:

{
  "status": "healthy",
  "timestamp": "2024-01-01T00:00:00.000Z",
  "service": "Article Generator API",
  "availableModels": 8
}

Project Structure

src/
├── config/
│   └── database.ts          # MongoDB connection
├── middleware/
│   └── errorHandler.ts      # Error handling middleware
├── models/
│   └── Article.ts           # Article data model
├── routes/
│   └── articles.ts          # API routes
├── services/
│   ├── ArticleGeneratorService.ts    # Main article generation logic
│   ├── EnhancedResearchService.ts    # Advanced web research with AI insights
│   ├── OpenRouterService.ts          # AI model integration
│   └── WebBrowsingService.ts         # Web content extraction and browsing
├── types/
│   ├── article.ts          # Article-related type definitions
│   └── api.ts              # API-related type definitions
├── utils/
│   └── helpers.ts          # Utility functions
└── index.ts                # Application entry point

Configuration

Environment Variables

Variable	Description	Default
OPENROUTER_API_KEY	Your OpenRouter API key	Required
MONGODB_URI	MongoDB connection string	mongodb://localhost:27017/article-generator
PORT	Server port	3000
NODE_ENV	Environment mode	development
DEBUG	Enable debug logging (1 or 0)	0
OPENROUTER_BASE_URL	OpenRouter API base URL	https://openrouter.ai/api/v1
DEFAULT_MODEL	Default AI model	openai/gpt-4o-mini
AVAILABLE_MODELS	Comma-separated list of models	See current model list below
RATE_LIMIT_WINDOW_MS	Rate limit window (milliseconds)	900000 (15 minutes)
RATE_LIMIT_MAX_REQUESTS	Max requests per window	100

Available AI Models

The following models are currently supported (as of default configuration):

• openai/gpt-4o-mini (default)
• google/gemini-2.5-flash
• anthropic/claude-sonnet-4
• anthropic/claude-3.7-sonnet
• google/gemini-2.5-pro
• x-ai/grok-4-fast
• openai/gpt-4.1-mini

You can customize available models by setting the AVAILABLE_MODELS environment variable with a comma-separated list.

Article Styles

• Feature: In-depth, magazine-style articles
• Breaking News: Fast, factual reporting
• Investigative: Deep-dive analysis with multiple sources
• Opinion: Editorial and commentary pieces
• Interview: Q&A format articles
• Profile: Biographical or company profiles

Article Tones

• Formal: Professional, business-like tone
• Conversational: Friendly, accessible language
• Authoritative: Expert, confident voice
• Casual: Relaxed, informal style
• Academic: Scholarly, research-oriented
• Humorous: Light-hearted, entertaining

Article Lengths

• Short: 300-600 words
• Medium: 600-1200 words
• Long: 1200-2500 words
• Long-form: 2500+ words

API Integration

The application integrates with OpenRouter API to access multiple AI models. You'll need to:

1. Sign up at OpenRouter
2. Get your API key
3. Add credits to your account
4. Configure the API key in your .env file

Development

Type Checking

bun run type-check

File Structure

• TypeScript: Strict typing enabled
• ESM: Modern ES modules
• Path Mapping: Convenient imports with @/ prefix
• Error Handling: Comprehensive error management
• Validation: Request validation with Zod

Contributing

1. Fork the repository
2. Create a feature branch
3. Make your changes
4. Add tests if applicable
5. Submit a pull request

Deployment

Using Docker (Recommended)

FROM oven/bun:latest

WORKDIR /app
COPY package.json bun.lockb ./
RUN bun install --frozen-lockfile

COPY . .
RUN bun run build

EXPOSE 3000
CMD ["bun", "start"]

Manual Deployment

1. Set up MongoDB on your server
2. Configure environment variables
3. Build the application: bun run build
4. Start the application: bun start

Monitoring and Logging

The application includes:

• Request logging
• Error tracking
• Performance monitoring
• Health check endpoints
• Rate limiting
• CORS configuration
• Security headers (Helmet)

Troubleshooting

Common Issues

1. MongoDB Connection Failed: Check your MongoDB URI and ensure MongoDB is running
2. OpenRouter API Errors: Verify your API key and account credits
3. Rate Limiting: Reduce request frequency or increase limits
4. Memory Issues: For long articles, consider increasing Node.js memory limit

Logs

Check console output for detailed error messages and debugging information.

License

MIT License - see LICENSE file for details.

Support

For issues and questions:

1. Check the troubleshooting section
2. Review the API documentation
3. Create an issue on GitHub
4. Contact support team

---

Built with ❤️ using Bun, TypeScript, Express, and MongoDB.