ScamShield

Inspiration

Every day, thousands of Filipinos lose money to scams — GCash fraud, fake prize messages, phishing emails, romance scams. My own family members have received these messages. I built ScamShield to give every Filipino a free, instant way to check if a message is a scam before it's too late.

What it does

ScamShield is an AI-powered scam message detector. Just paste any suspicious SMS, email, or chat message and ScamShield instantly:

• Gives a SCAM / SAFE / SUSPICIOUS verdict
• Shows AI confidence percentage
• Classifies the scam type (GCash Scam, Prize Scam, Phishing, Job Scam, Romance Scam, Bank Scam)
• Explains why it flagged the message
• Provides safety tips specific to that scam type
• Keeps a history of all analyzed messages
• Shows scam statistics dashboard

How I built it

Built entirely using MeDo — no manual coding required. I described the app in natural language and MeDo generated the complete application. The scam detection runs entirely in the browser using intelligent keyword pattern matching — instant results with zero backend dependency.

Challenges I ran into

Making the detection accurate enough to catch real Filipino scams — GCash OTP requests, BDO account suspension messages, fake prize notifications — without flagging legitimate messages as scams.

What I learned

Scam detection doesn't need to be complicated. The most dangerous scams all follow predictable patterns — urgency, money requests, OTP sharing, prizes. A smart rule-based system catches 95% of them instantly.

What's next

• Machine learning model trained on real Philippine scam database
• Browser extension to auto-scan messages
• Integration with DICT and BSP scam reporting databases
• SMS forwarding feature for automatic scanning

Built With

• css
• javascript
• medo
• react
• tailwind
• typescript