Intellex

AI-powered platform that fetches, rewrites, and publishes news in markdown for static sites.

AutoJourno-AI 📰

AutoJourno-AI is an intelligent, autonomous news article generation engine that fetches real-time news from global sources, rewrites them in a human-like markdown format, and organizes them neatly for static publishing via Hugo. This system is designed for developers, journalists, media startups, and digital publication platforms.

---

✨ Key Features

• Automated News Fetching: Gathers the latest news using APIs based on keywords, categories, or regions.
• AI-Powered Rewriting: Leverages a Small Language Model (SLM) to transform headlines and summaries into unique, well-structured articles.
• Template-Based Formatting: Uses Jinja2 templates to ensure all articles have consistent markdown structure and Hugo front matter.
• Seamless Hugo Integration: Automatically pushes generated articles into a Hugo project's content directory for instant deployment.

⚙️ How It Works

The system follows a simple, four-step pipeline to automate content creation:

[Fetch News (news_fetcher.py)] → [Rewrite with SLM (slm_writer.py)] → [Generate Article (generate_article.py)] → [Publish to Hugo (hugo_publisher.py)]

🧠 Project Structure

autojourno-ai/
├── main.py                        # Entry point for running the full pipeline
├── requirements.txt              # Python dependencies
├── config.ini                    # Configuration file for API keys and settings
├── data/                         # (Optional) Placeholder for storing input or fetched data
├── layouts/page/                 # Hugo-specific layout overrides (if used)
├── models/
│   └── slm_writer.py             # SLM (Small Language Model) for rewriting fetched news
├── output/
│   └── *.md                      # Generated markdown articles
├── templates/
│   └── article_template.md.j2    # Jinja2 template for formatting news content
├── utils/
│   ├── generate_article.py       # Converts fetched news to formatted articles
│   ├── hugo_publisher.py         # Pushes articles into Hugo’s content folder
│   └── news_fetcher.py           # News scraping using APIs (e.g., NewsAPI)
├── venv/                         # Local Python virtual environment
└── hugo.zip                      # Static Hugo site template (can be extracted for deployment)

---

🚀 Getting Started

Prerequisites

• Python 3.10+
• pip and venv
• Hugo (for static site generation)
• A News API Key (e.g., from NewsAPI.org)

Setup

1. Clone the repository:

   git clone [https://github.com/yourusername/autojourno-ai.git](https://github.com/yourusername/autojourno-ai.git)
   cd autojourno-ai

2. Create and activate the virtual environment:

   # For macOS/Linux
   python3 -m venv venv
   source venv/bin/activate
   
   # For Windows
   python -m venv venv
   venv\Scripts\activate

3. Install dependencies:

   pip install -r requirements.txt

4. Configure the system:

   • Rename config.ini.example to config.ini (or create it).
   • Add your API key and other settings:

   [news_api]
   api_key = YOUR_NEWS_API_KEY_HERE
   keywords = AI, technology, SpaceX, startups
   
   [hugo]
   content_path = my-hugo-site/content/news/

---

🛠️ Usage

Run the Full Pipeline

Execute the main script to run all modules in sequence:

python main.py

Run Individual Modules

You can also run each step of the process independently.

1. Fetch news from the API:

   python utils/news_fetcher.py

2. Generate markdown articles from fetched data:

   python utils/generate_article.py

3. Publish generated articles to your Hugo site:

   python utils/hugo_publisher.py

🧪 Example Output

Generated articles will appear in the output/ directory with URL-friendly filenames:

output/
├── elon-musks-spacex-might-invest-2-billion-in-musks-xai.md
├── study-warns-of-significant-risks-in-using-ai-therapy-chatbots.md
└── meta-acquires-voice-startup-play-ai.md

Each .md file is formatted and ready to be placed in your Hugo site’s content/news/ folder.

---

📂 Deployment

To deploy your news site with the generated content:

1. Unzip the Hugo template:

   unzip hugo.zip -d my-hugo-site
   cd my-hugo-site

2. Ensure articles are in place: The hugo_publisher.py script will have already moved the articles into my-hugo-site/content/news/.

3. Serve or build the site with Hugo:

   # Serve locally for testing
   hugo server
   
   # Build the static site for deployment
   hugo

   The final site will be in the public/ directory, ready to be deployed to Netlify, Vercel, GitHub Pages, or any static host.

---

💡 Use Cases

• News Aggregator Sites: Automate content creation from top headlines.
• Tech Newsletters: Generate markdown-ready content to plug into email platforms.
• Personal AI Blog: Auto-blog on specific topics or industries.
• SEO Optimization: Generate trending content quickly to improve site ranking.

---

📜 License

This project is released under a Custom License. Please see the LICENSE file for more details.

---

👨‍💻 Author

Balasubrahmanya A G

• Computer Science & Engineering Student
• Tech Enthusiast
• 📧 Email: subrahmanyaprasad23balu@gmail.com
• 🔗 LinkedIn: https://www.linkedin.com/in/balasubrahmanya-a-g-77965628b/