CyberSafe AI Coach

AI-powered scam detection web app that analyzes suspicious messages, assigns risk scores, and educates user with real-world cyber safety guidance.

Comment

Inspiration : After eight and a half years in the healthcare industry, i saw firsthand how essential automated precision is to safety. As cyber scams and phishing attacks grow more sophisticated, i realized that everyday user need that same level of automated protection. Many struggle to spot red flags like artificial urgency or fake threats. I developed this app to bridge that gap creating a practical tool that doesn't just detect scams, but actively educates users on how to stay safe.

What it does: CyberSafe AI Coach is a web-based scam detection application that analyzes suspicious SMS, email, or text messages and a risk score from 0-100.

The system identifies scam indicators such as:

  • Request for sensitive information (PIN, OPT, passwords)
  • Financial keywords (ATM, bank, account)
  • Urgency and threat language The application then provides:
  • Risk level (Low/Medium/High)
  • Confidence percentage
  • Explanation of detected scam signals
  • Cyber safety advice

Additional features include:

  • Screenshot upload support
  • Live screen capture support
  • Fake vs Safe comparison mode

How we built it: The project was built using:

  • HTML
  • CSS
  • JavaScript I designed a custom rule-based detection engine that stimulates AI- style reasoning while remaining lightweight and fully offline. The system is also designed to support future real-time AI API integration for more advance scam detection capabilities. The UI was built to be clean, responsive, and beginner- friendly.

Challenges we ran into: One challenge was improving scam detection accuracy while keeping the system fully offline and lightweight. Another challenge was creating a clean and interactive user experience within a short hackathon timeframe.

I also worked on balancing usability with cybersecurity concepts so non-technical users could easily understand the results.

Accomplishments that we're proud of :

  • Built a complete working prototype
  • Designed a modern cybersecurity themed UI
  • Added screenshot and screen capture functionality
  • Created an educational cyber awareness experience

What we learned:

Through this project, i learned more about:

  • Scam and phishing detection patters
  • Fronted UI/UX design
  • Browser media APIs
  • Real-world cybersecurity awareness challenges I also learned how important user education is in cybersecurity solutions.

What's next for CyberSafe AI Coach:

Future improvements include:

  • AI API integration for advance scam analysis
  • OCR test extraction from screenshots
  • Browser extension support
  • Mobile application version
  • Real-time email and sms scanning

Built With

Share this project:

Updates

posted an update —

Excited to submit CyberSafe AI Coach to the DevNetwork [AI + ML] Hackathon 2026 CyberSafe AI Coach is a web-based scam detection tool that analyzes suspicious messages, generates risk scores, and helps users about phishing and cyber threats through an interactive and beginner-friendly experience. Key features include:

  • Risk scoring system
  • Scam signal explanations
  • Screenshot upload support
  • Live screen capture support
  • Fake vs Safe comparison mode Built using HTML, CSS, and JavaScript.

Log in or sign up for Devpost to join the conversation.