- Make sure you have Docker installed and running
- Clone this repo
- Install dependencies, including the Supabase CLI
yarn
Note: If you install the Supabase CLI using a different method you have to make sure you are on version 1.49.4 as more recent versions currently suffer from an issue which prevents this from working correctly.
- Create frontend env file
cp .env.example .env
- Create supabase functions env file
echo "OPENAI_API_KEY=sk-xxx" > supabase/.env
- If not already running, start Docker. Learn how to do this for your OS here.
- Start the supabase project.
npx supabase start
- Start the supabase functions locally
yarn supabase:dev
- Start the frontend locally
yarn dev
- Open http://localhost:3100 with your browser to see the result.
- Create a new project on Supabase
- Create a new project on Vercel
- To deploy the frontend, connect your Vercel project to your GitHub repo and push to main.
- To deploy the supabase functions, first login to Supabase:
npx supabase login
Then, link your project:
npx supabase link --project-ref <project-ref>
Then, deploy the functions:
yarn supabase:deploy
Optionally, if you're also using the Supabase Vector Store from LangcChain, you'll need to push the schema to the database:
supabase db push