How keeganjustis/wa-state-environmental-policy-watch works
Project type and purpose
• This is a prototype Python-based backend service (“Environmental Policy Newsfeed & AI Summarization Tool”) that periodically fetches environmental-policy articles (focused on Washington State Legislature), summarizes them via an AI model, and (in the future) will store results and serve/push notifications.
File structure
• Top-level: CI workflows (.github/workflows/pylint.yml), config files (requirements.txt, .gitignore), tests (test/unit).
• src/: main orchestration (main.py), three pipeline stages: get_news.py (fetching), analyze_news.py (summarization), send_summary.py (delivery).
• tmp/news_sites.json: list of source endpoints.
README for insights
• Core pipeline: Source Identification → Article Retrieval → AI Summarization → (future) Storage & Notification → (planned) Frontend/API.
• Technology stack: Python, LLM (e.g., GPT API), future DB (PostgreSQL), future web/mobile UI, notifications via email/SMS.