AI Chatbot

A full-featured, hackable AI chatbot with authentication, file storage, and multi-model support.

Tech Stack

• Framework: Next.js 15
• React: React 19
• AI: AI SDK, Vercel AI Gateway
• Auth: Auth.js
• ORM: Drizzle
• Database: PostgreSQL (Neon, Supabase, or Railway)
• Styling: Tailwind CSS, shadcn/ui
• Package Manager: pnpm
• Dev Port: 3000

Setup

1. Clone the Template

git clone --depth 1 https://github.com/Eng0AI/ai-chatbot-template.git .

If the directory is not empty:

git clone --depth 1 https://github.com/Eng0AI/ai-chatbot-template.git _temp_template
mv _temp_template/* _temp_template/.* . 2>/dev/null || true
rm -rf _temp_template

2. Remove Git History (Optional)

rm -rf .git
git init

3. Install Dependencies

pnpm install

4. Setup Environment Variables

cp .env.example .env

Required variables:

• POSTGRES_URL - PostgreSQL connection string
• AUTH_SECRET - Generate with openssl rand -base64 32
• OPENAI_API_KEY or ANTHROPIC_API_KEY - LLM provider key

5. Run Database Migrations

pnpm db:migrate

Build

pnpm build

Or run build without migration (if already migrated):

next build

Deploy

Vercel (Recommended)

# Pull project settings
vercel pull --yes -t $VERCEL_TOKEN

# Push env vars (first time only)
while IFS='=' read -r key value; do
  [[ "$key" =~ ^#.*$ || -z "$key" || -z "$value" ]] && continue
  for env in production preview development; do
    printf '%s' "$value" | vercel env add "$key" $env -t $VERCEL_TOKEN
  done
done < .env

# Build and deploy
vercel build --prod -t $VERCEL_TOKEN
vercel deploy --prebuilt --prod --yes -t $VERCEL_TOKEN

Netlify

# Import env vars (first time only)
netlify env:import .env

# Deploy
netlify deploy --prod

Critical Notes

• Database Required: Must have PostgreSQL database set up before building
• Migration Required: Run pnpm db:migrate before first build
• Auth Secret: Generate a secure random secret for AUTH_SECRET
• Never run pnpm dev in VM environment