PromptVault

• 

  Landing page

• 

  Example prompt

Inspiration

I kept losing good AI prompts — they'd end up in random notes, tweets, or just forgotten. I wanted one place where the best prompts could live and be discovered by everyone, not just saved privately by one person.

What it does

PromptVault lets users submit, browse, search, and upvote AI prompts organized by category (Coding, Writing, Image Generation). The community surfaces the best prompts through upvotes, so quality rises to the top naturally.

How I built it

Built with Next.js and Tailwind CSS on the frontend, Supabase for the database and authentication, and deployed on Vercel. The upvoting logic is enforced at the database level using PostgreSQL constraints, and authentication uses secure email-based flows with open-redirect protection.

Challenges we ran into

Getting from local development to production was harder than expected. A misconfigured framework preset on Vercel caused site-wide 404 errors, and Supabase auth redirect URLs needed careful configuration for the live domain. Debugging production issues without breaking what already worked locally was the real test.

Accomplishments that we're proud of

Shipping a fully functional, production-grade app with real authentication, a live upvoting system, and security hardening — all working cleanly on a public URL.

What we learned

That the gap between "works on my machine" and "works in production" is real. Also how much value there is in enforcing data integrity at the database level rather than relying purely on application code.

What's next for Prompt Vault

Adding user profiles, prompt collections, and tagging. Eventually, integrating an AI layer that suggests similar prompts or auto-categorizes new submissions — making discovery even smarter.

Built With

• next.js
• postgresql
• react
• supabase
• tailwindcss
• typescript
• vercel