README.md

---
title: DisasterTracker AI 🌍
emoji: 🌪️
colorFrom: blue
colorTo: green
sdk: streamlit
sdk_version: 1.44.1
app_file: app.py
pinned: false
---

# 🌍 DisasterTracker AI

**DisasterTracker AI** is a fullstack open-source RAG app that helps you stay updated on recent natural disasters around the world. It combines the power of local embeddings, open-source language models, real-time news, and live weather data.

## 🔍 Features

- ✅ **Retrieval-Augmented Generation (RAG)** with `flan-t5-base`
- ✅ **Local vector store** using ChromaDB and SentenceTransformers
- ✅ **Live RSS feeds** for global natural disaster news
- ✅ **Real-time weather forecasts** from Open-Meteo API
- ✅ **Streamlit UI** optimized for Hugging Face Spaces (no API keys required)
- ✅ Runs entirely on **CPU** — 100% open source and deployable for free

## 🚀 How It Works

1. **News** is scraped from RSS feeds (NYT, Reuters, Al Jazeera).
2. Articles are **embedded** using `all-MiniLM-L6-v2` and stored in **ChromaDB**.
3. When you ask a question, the system:
   - Finds relevant articles (RAG)
   - Generates an answer using `flan-t5-base`
   - Fetches live **weather** for mentioned locations

## 🛠️ Setup Instructions

### 1. Install Dependencies
```bash
pip install -r requirements.txt