Hrittik-Chatterjee/scamshield — reverse-engineered prompt

Reverse engineered prompt

I want you to turn this starter Astro project into a real app called ScamShield that helps everyday people figure out if a text, email, or chat message might be a scam. Make it feel polished and trustworthy, not like a template. The core experience should be simple, paste in suspicious text or upload a chat screenshot, then show an easy to understand result with the warning signs, why it looks risky or safe, and what the person should do next.

Please use any notes, mockups, or docs already in the repo if they exist, and keep the design clean and modern. Add the pages and flows needed so it feels like a complete product, including a clear homepage and a usable checking flow. If there is already some database or deploy setup in here, wire it up in a sensible way. I should be able to run it locally and see the full experience. Look up current docs online if you need to.

Want more depth? Deep Reverse