A Python program to fetch recently published climate research articles and summarize them. Generates weekly or monthly digests of top climate articles from journals and blogs.
- Fetches articles from major climate science journals:
- Nature, Nature Climate Change, Nature Geoscience, Nature Communications, Science, Science Advances, Journal of Climate, Climate Dynamics, Geophysical Research Letters, JGR: Atmospheres & Oceans, Earth’s Future, Atmospheric Research, and more.
- Fetches articles from blogs & news sources:
- Skeptical Science, RealClimate, NOAA Climate.gov, Grantham Institute, The Economist (Climate), The Conversation (Climate), Climate Brink.
- Filters articles to climate-related topics using keywords.
- Selects top 5–10 most important articles based on source weights and recency.
- Summarizes articles using Sumy extractive summarization:
- Uses RSS summary + title for all articles.
- Optionally, top articles can be summarized from full text for richer highlights.
- Produces a Markdown digest with:
- Top 3 highlights section
- Grouped journals and blogs
- Author, publication date, and RSS feed link
- Lightweight, runs without any paid APIs.
- Configurable for weekly or monthly digests.
- Clone the repository:
git clone https://github.com/yourusername/research_article_fetcher.git
cd research_article_fetcher- Install dependencies
pip install -r requirements.txt- Run the script.
python main.py- Choose digest frequency in main.py
# For weekly digest (default)
cutoff_days = 7
# For monthly digest
cutoff_days = 30- Output markdown digest is saed in the
output/folder
output/climate_digest.mdresearch_article_fetcher/
├─ main.py # Main script to generate digest
├─ fetch/
│ ├─ feeds.py # RSS feed URLs
│ ├─ rss_reader.py # Fetches articles from feeds
│ └─ filter.py # Ranking and filtering logic
├─ summarizer.py # Summarizes articles using Sumy
├─ utils/
│ └─ helpers.py # Markdown formatting helpers
├─ output/ # Generated Markdown digests
└─ requirements.txt # Python dependencies- Fetch full article text for richer summarization of top articles.
- Add PDF export in addition to Markdown.
- Add highlight emojis/icons for top articles or categories.
- Expand source feeds to cover more climate journals/news outlets.
- Add keyword-based filtering for specific topics (ENSO, monsoon, sea-level rise, etc.).
- Implement automatic scheduling for weekly/monthly digest generation.
- Optional interactive web version for displaying digests.
- Improve summarization quality using hybrid methods for top articles.