Hugging Face's logo

Spaces:
ztcoco
/
Smart-Web-Monitor
Paused

App Files Community
Smart-Web-Monitor / README.md
WebAI Deployer
Update Camouflage App (2026-01-10)
b36d0b3
preview code
|
raw
history blame contribute delete
2.54 kB
metadata 
title: Smart Web Monitor
emoji: 🔍
colorFrom: blue
colorTo: purple
sdk: docker
pinned: false
app_port: 7860

🔍 Smart Web Monitor

AI-Powered Website Change Detection System

Monitor websites for changes automatically with AI-driven content analysis. Perfect for tracking competitor updates, news sites, or any web content you care about.

✨ Features

  • 🕐 Automated Monitoring: Check websites every 5 minutes automatically
  • 📸 Content Hash Detection: Track changes via MD5 hash comparison
  • 🤖 AI Sentiment Analysis: Powered by DistilBERT model from HuggingFace
  • 🔍 Manual Checks: Instant verification anytime
  • 📊 History Tracking: Review all past checks
  • 🎯 Multi-URL Support: Monitor unlimited websites

🤖 AI Technology

This project uses real HuggingFace Transformers:

  • Model: distilbert-base-uncased-finetuned-sst-2-english
  • Task: Sentiment Analysis (POSITIVE/NEGATIVE classification)
  • Purpose: Detect tone changes in web content over time

🚀 Quick Start

  1. Add URLs: Go to "Monitor Management" tab and add websites
  2. Auto-Check: System automatically checks every 5 minutes
  3. Manual Check: Use "Manual Check" tab for instant verification
  4. View History: Check "History" tab to see all results

📋 Use Cases

  • 📰 News monitoring
  • 🏢 Competitor tracking
  • 💰 Price change alerts
  • 📝 Content update detection
  • 🔔 Government notice tracking

🛠️ Technology Stack

  • Frontend: Gradio 4.x
  • Backend: Python 3.11
  • Browser Engine: Chromium (for advanced scraping)
  • Deployment: HuggingFace Spaces (Docker SDK)

⚙️ Configuration

Set these environment variables in HuggingFace Spaces settings:

# Optional: Custom check interval (default: 5 minutes)
CHECK_INTERVAL=300

# Optional: Maximum URLs to monitor (default: 50)
MAX_URLS=50

📊 How It Works

  1. Hash-Based Detection: Each check computes MD5 hash of page content
  2. Background Worker: Daemon thread runs checks every 5 minutes
  3. History Persistence: Recent 100 checks stored in memory
  4. Smart Scheduling: Distributed checks to avoid rate limiting

🔒 Privacy

  • No data is stored permanently
  • All checks run in-memory
  • No external logging or tracking
  • Your URLs stay private

📝 License

MIT License - Free for personal and commercial use

🤝 Contributing

Found a bug? Have a feature request? Open an issue on GitHub!

Made with ❤️ by the HuggingFace Community