Fraud Shield

• 

  Extension Logo

Inspiration

Scams and phishing emails are on the rise. We wanted to create a tool that uses AI to help users spot and avoid fraud in real time.

What it does

Fraud Shield scans emails and messages for suspicious content, uses AI to evaluate fraud risk, and alerts users with a confidence score and risk reasons.

How we built it

We used JavaScript and Chrome APIs to create a browser extension. For fraud detection, we integrated a generative AI model through Minstral's API to analyze content.

Challenges we ran into

• Parsing email content accurately in a browser
  
• Getting consistent API responses
  
• Balancing detection accuracy with false positives
  
• Understanding limitations of free API usage

Accomplishments that we're proud of

• Built a working prototype with real-time scanning
  
• Successfully integrated AI fraud detection
  
• Created a clean and simple UI for alerts
  
• Maintained strong performance in the browser

What we learned

• How to work with Chrome extensions and local storage
  
• Prompt engineering for AI-based classification
  
• Handling asynchronous messaging in browser scripts
  
• The importance of user trust and clear warnings

What's next for Fraud Shield

• Train a custom AI model on real scam data
  
• Run locally instead of an API call

Built With

• css
• html
• javascript
• mistral