ScamVerifier / README.md
AayanSuleri's picture
Update README.md
6da26d2 verified
|
Raw
History Blame Contribute Delete
4.75 kB

A newer version of the Gradio SDK is available: 6.19.0

Upgrade
metadata
license: mit
title: Scam Verifier
sdk: gradio
emoji: πŸ’»
colorFrom: blue
colorTo: red
short_description: Verify wether the message is a scam or not
sdk_version: 5.42.0

πŸ›‘οΈ Scam-Signal Verifier

A comprehensive web application that helps users identify potentially fraudulent messages, emails, and advertisements. Perfect for protecting students, elderly individuals, and anyone who wants to stay safe from online scams.

🌟 Features

Multi-Agent Analysis System

  • Claim Extractor: Parses messages for URLs, phone numbers, emails, and financial amounts
  • Pattern Verifier: Analyzes content against known scam patterns and suspicious keywords
  • Risk Calculator: Provides a comprehensive risk score (0-100) with clear classification
  • Human-Friendly Explainer: Generates actionable recommendations and reporting templates

Detection Capabilities

  • Phishing Emails: Bank, PayPal, Amazon verification scams
  • Job Offer Scams: Work-from-home and "easy money" schemes
  • Investment Fraud: Cryptocurrency and get-rich-quick schemes
  • Lottery/Prize Scams: Fake winnings and inheritance claims
  • Urgency Tactics: High-pressure language and artificial deadlines
  • Suspicious Domains: Known malicious and shortened URLs

User Experience

  • Clean, modern interface with gradient design
  • Real-time risk assessment with color-coded results
  • Pre-loaded example scams for testing
  • One-click report template generation
  • Mobile-responsive design

πŸš€ Quick Start

Local Development

# Clone the repository
git clone <repository-url>
cd scam-signal-verifier

# Install dependencies
pip install -r requirements.txt

# Run the application
python app.py

Docker Deployment

# Build the image
docker build -t scam-verifier .

# Run the container
docker run -p 7860:7860 scam-verifier

Hugging Face Spaces

This app is optimized for Hugging Face Spaces deployment:

  1. Upload all files to your Spaces repository
  2. Set the Space to use Python/Gradio environment
  3. The app will automatically start on port 7860

🎯 How It Works

1. Message Analysis

The system extracts key information from suspicious messages:

  • URLs and domain reputation
  • Contact information (phone, email)
  • Financial references and amounts
  • Suspicious keywords and phrases
  • Urgency indicators and pressure tactics

2. Risk Assessment

Multi-factor scoring algorithm considers:

  • Keyword Score (0-40 points): Suspicious language patterns
  • Urgency Score (0-25 points): High-pressure tactics
  • Domain Risk (0-20 points): Suspicious or shortened URLs
  • Contact Risk (0-10 points): Multiple contact methods
  • Financial Risk (0-15 points): Money-related content

3. Classification System

  • πŸ”΄ HIGH RISK (80-100): Obvious scam, avoid completely
  • 🟑 MEDIUM RISK (50-79): Suspicious, verify independently
  • 🟒 LOW RISK (25-49): Minor concerns, stay cautious
  • βœ… MINIMAL RISK (0-24): Appears legitimate

πŸ“Š Example Results

High-Risk Phishing Email

Risk Score: 95/100 - HIGH RISK
Red Flags:
🚨 Contains 6 suspicious keywords/phrases
🚨 Uses high-pressure urgency tactics  
🚨 Uses suspicious domain: secure-bank-verify.net
🚨 Mentions money/financial transactions

Recommendations:
🚫 Do NOT click any links or provide personal information
πŸ“§ Report this message as spam/phishing
πŸ›‘οΈ Block the sender immediately

πŸ›‘οΈ Safety Features

  • No External API Calls: All analysis happens locally for privacy
  • Educational Focus: Teaches users to recognize scam patterns
  • Report Generation: Creates templates for reporting to authorities
  • Multiple Examples: Pre-loaded scenarios for learning

πŸ”§ Technical Stack

  • Backend: Python Flask with modular analysis engine
  • Frontend: Responsive HTML/CSS/JavaScript
  • Deployment: Docker + Gunicorn for production
  • Security: No data persistence, privacy-focused design

πŸ“± Mobile Support

The interface is fully responsive and works seamlessly on:

  • Desktop browsers
  • Mobile phones
  • Tablets
  • Touch devices

🀝 Contributing

This is an educational tool designed to help people stay safe online. Feel free to:

  • Report bugs or issues
  • Suggest new scam patterns to detect
  • Improve the UI/UX design
  • Add new analysis features

πŸ“„ License

Open source educational project. Use responsibly to help others stay safe from online scams.

⚠️ Disclaimer

This tool is for educational purposes and general guidance. It should not be the sole method for determining if something is a scam. Always use your best judgment and consult with relevant authorities when needed.


Stay Safe Online! πŸ›‘οΈ