.png&w=48&q=75){kind=small}
OrpehusAthon
Goal: Access curated news easily (made for my mom & me). Stack: Python, Gemini API, GitHub Actions, Resend, feedparser. Time: ~1 hour. Idea: My mom asked for an easy way to read relevant news without clutter, so I automated a serverless AI curator that sends an HTML digest every Monday.
- 1h
- Coding
- Python
- Top language
📰 AI News Curator Agent
An autonomous, lightweight, and 100% free AI agent for smart content curation. It fetches articles from your favorite RSS feeds, filters out clickbait and noise based on your personal interest profile using Gemini 2.5 Flash, and sends a beautifully formatted HTML briefing directly to your inbox every Monday morning.
Runs completely serverless in the cloud—no need to keep your PC on or pay for hosting!
Tech Stack
- Language: Python 3.10
- AI / LLM: Google Gemini API (
google-genai) - Automation / CI/CD: GitHub Actions (Cron Jobs)
- Email Delivery: Resend API
- Data Ingestion:
feedparser
How It Works
- Scheduled Trigger (Cron): Every Monday at 07:00 AM (BRT), GitHub Actions triggers the script execution.
- RSS Collection: The agent scrapes configured RSS feeds for recent posts.
- Smart Curation: Gemini evaluates article titles and summaries against your interest profile to discard irrelevant corporate noise and clickbaits.
- Email Dispatch: A styled HTML report is generated and sent directly to your email inbox via Resend.
Getting Started
- Clone the repository:
git clone [https://github.com/danielmcneto/ai-news-scout.git](https://github.com/danielmcneto/ai-news-scout.git) cd ai-news-scout - Install dependencies:
pip install -r requirements.txt - Create a .env file in the root directory:
GEMINI_API_KEY=your_gemini_api_key RESEND_API_KEY=your_resend_api_key MY_EMAIL=your_email@example.com - Set up Repository Secrets in GitHub (Settings > Secrets and variables > Actions):
- GEMINI_API_KEY
- RESEND_API_KEY
- MY_EMAIL