This Python project fetches the latest Google News RSS headlines for a user-supplied list of topics and regularly updates a static HTML page. It uses Google Gemini to prioritize headlines and also to update a global outlook section biweekly. Designed to be scheduled with cron on any Unix-based system.
How It Works
- History Tracking
Sent articles are stored locally to prevent duplicates in future emails. - Fetch Articles
For each user-defined topic, NewsBot pulls recent headlines using Google News RSS. - Filter Old Content
Headlines that have appeared in recent digests are skipped to avoid repetition. - User Preferences
The user provides a list of topics and keywords, each with a score from 1 to 5 indicating importance. There’s also support for banned or demoted topics. - Prioritization with Gemini
Headlines are passed to Gemini, which selects a small set of topics and headlines based on the user’s ranked preferences.