faisal-34/gandabot

Luganda language learning app with AI tutor, clan lokup, flashcards, and Buganda cultural content # GandaBot — AI-Powered Luganda Learning Next.js 15 + Clerk + Neon PostgreSQL, deployed on Vercel. ## Quick Start ### 1. Clone & install ```bash git clone github.com cd gandabot npm install ``` ### 2. Set up environment variables ```bash cp .env.example .env.local # Fill in all required values (see below) ``` ### 3. Set up database ```bash node scripts/migrate.js ``` ### 4. Run locally ```bash npm run dev # Open localhost ``` --- ## Deploy to Vercel ### Option A — Vercel Dashboard (recommended) 1. Push repo to GitHub 2. Go to vercel.com → Import your repo 3. Framework: **Next.js** (auto-detected) 4. Add all environment variables from `.env.example` 5. Click Deploy ### Option B — CLI ```bash npm i -g vercel vercel --prod ``` --- ## Environment Variables | Variable | Required | Description | |---|---|---| | `NEXT_PUBLIC_CLERK_PUBLISHABLE_KEY` | ✅ | Clerk public key | | `CLERK_SECRET_KEY` | ✅ | Clerk secret key | | `DATABASE_URL` | ✅ | Neon PostgreSQL connection string | | `NEXT_PUBLIC_APP_URL` | ✅ | Production URL (`gandabot.com`) | | `OPENAI_API` | Recommended | GPT-4o-mini for chat, moderation, insights | | `ELEVENLABS_API_KEY` | Optional | Podcast text-to-speech audio generation | | `RESEND_API_KEY` | Optional | Welcome + notification emails | | `NEXT_PUBLIC_GOOGLE_MAPS_API_KEY` | Optional | Country Explorer live places data | --- ## Clerk Setup 1. Create account at clerk.com 2. Create a new application 3. Enable **Google** and **Email/Password** sign-in methods 4. Copy `NEXT_PUBLIC_CLERK_PUBLISHABLE_KEY` and `CLERK_SECRET_KEY` 5. In Clerk dashboard → **Paths**: set sign-in to `/sign-in`, sign-up to `/sign-up` --- ## Database (Neon) 1. Create account at neon.tech 2. Create a new project 3. Copy the connection string to `DATABASE_URL` 4. Run `node scripts/migrate.js` to create all tables --- ## Features | Feature | Route | Description | |---|---|---| | Landing Page | `/` | Marketing page …