Inspiration

This web app was inspired by real events in my life. My mother was nearly scammed by someone posing as a bank official and had already shared two out of three OTPs before we realized something was wrong. We narrowly avoided the scam because the requests for OTPs seemed suspicious. This experience motivated me to build a web app that helps not just my family, but anyone, to recognize scam attempts and protect themselves from fraud.

What it does

SuusyBaka is a privacy-first scam detection tool that analyzes suspicious texts, audio messages, and images in seconds. It detects 15+ scam patterns, generates a clear risk score, explains why something looks suspicious, and tells users exactly what to do next—without sending any data off the device.

How I built it

We built SuusyBaka using TypeScript and React, with a custom pattern-matching engine and weighted risk-scoring system. Everything runs 100% client-side in the browser, including text analysis, audio transcription, and image text extraction. Accessibility and mobile usability were core design priorities from day one.

Challenges we ran into

Balancing accuracy with speed while keeping all processing local was tough. Designing explanations that are clear for non-technical users—especially seniors—without being overwhelming was another major challenge. Accessibility compliance also required careful iteration and testing.

Accomplishments that we're proud of

Fully on-device scam detection (no servers, no tracking) Support for text, audio, and image-based scams 15+ scam categories with actionable explanations WCAG 2.1 AA accessibility and mobile optimization Free, open-source, and usable offline

What we learned

Scam prevention isn’t just about detection—it’s about education and confidence. Users feel safer when they understand why something is dangerous and what steps to take next. Privacy and simplicity dramatically increase trust and adoption.

What's next for SuusyBaka

Expanded scam pattern library (new and emerging tactics) Browser extensions and PWA support Improved educational content and real-time scam alerts Localization for non-English users Partnerships with senior advocacy and consumer protection groups

Share this project:

Updates