Maurice-Muthaka/supabase-vectors-template

template-next-supabase

Starterkit for OpenAI chat client based on your own content. Main ingredients:

• Next.js
• Supabase + pgvector
• @7-docs/edge
• TypeScript
• Tailwind

Install

• Use this template
• Clone your new repo
• npm install
• Prepare vector database
• Ingest content
• cp .env.example .env.local and paste your tokens: OPENAI_API_KEY, SUPABASE_URL, SUPABASE_API_KEY

The following commands are using my-namespace as set in config.ts and ingest content from the reactjs/react.dev repository. By all means, you're encouraged to change them to something more original.

Prepare db

Create a Supabase account if you don't have one already.

The following command does not actually do anything. Instead, it gives you an all-batteries-included SQL query to run in your Supabase project area's SQL Editor:

npx 7d supabase-create-table --namespace my-namespace

Ingest

npx 7d ingest \
  --source github \
  --repo reactjs/react.dev \
  --files 'src/content/reference/react/*.md' \
  --files 'src/content/reference/react-dom/*.md' \
  --files 'src/content/learn/*.md' \
  --namespace my-namespace

Run

npm run dev

Deploy

This application can be deployed on platforms with support for Next.js + edge functions. Examples:

• Vercel
• Fly.io

Make sure to add the environment variables to the platform first.