CyberWhisper – Real-Time Scam and Threat Analyzer

CyberWhisper: Real-time URL and message threat detection, powered by AI and your voice.

Comment

Inspiration

Every day, millions of people encounter suspicious links or any message online—on social media, email, and messaging apps. Seeing friends and family fall victim to scams inspired me to create a tool that makes checking link or message safety as easy and accessible as possible, for everyone.

What it does

CyberWhisper is a real-time scam and threat analyzer for URLs or message. Users can paste or speak any link or message, and instantly get feedback on its safety—powered by the VirusTotal API and our intelligent backend. The app supports both traditional link scanning and a voice assistant, making it quick and intuitive to check any link or message on the go.

How we built it

I built CyberWhisper using:

  • React for the responsive frontend
  • Express.js for the backend API
  • VirusTotal API for link threat intelligence
  • Bolt.new for rapid prototyping and deployment
  • Netlify for public deployment
  • Voice assistant integration for both desktop and mobile web
  • My initial plan was to use the Algorand blockchain to store and verify link analysis results for added transparency. However, because of CORS issues and deployment limitations, we substituted the blockchain calls with a mock module that simulates the expected behavior.

I focused on a clean UX, robust API integration, and real-time feedback for users.

Challenges we ran into

  • Integrating the VirusTotal API with rate limits and key management
  • Handling CORS and security restrictions in both local and deployed environments
  • Voice assistant functionality across different browsers and devices
  • I attempted to integrate the Algorand blockchain for on-chain link validation and transparency. However, due to CORS restrictions on the available APIs, I was unable to make direct requests from my frontend/backend deployed on Bolt.new/Netlify. To keep the user experience seamless and to demonstrate the intended blockchain features, I implemented a mock version simulating the Algorand interactions.

Accomplishments that we're proud of

  • Full voice assistant support for hands-free link or message scanning
  • Successfully deploying a public demo with minimal setup
  • Creating a user-friendly, accessible interface
  • Real-time threat analysis with clear, actionable feedback
  • Overcoming API and deployment challenges

What we learned

  • Deepened my understanding of web security and threat intelligence
  • Improved skills in building cross-platform and voice-enabled apps
  • Learned to work around platform constraints (Bolt.new/Netlify/API keys)
  • Gained experience in user onboarding for security tools

What's next for CyberWhisper – Real-Time Scam and Threat Analyzer

  • Add support for scanning files and attachments
  • Integrate with more threat intelligence APIs
  • Build browser extensions for even easier use
  • Support more languages for global accessibility
  • Add user accounts for scan history and advanced features

Built With

Share this project:

Updates