Install Aurora Catalyst

With the CLI installed, this step scaffolds a new Aurora Catalyst project. The procedure is the same for any Catalyst app — the role it plays is decided later, from configuration.

1. Scaffold the project

catalyst new my-project

This creates an Aurora monorepo (a backend/ + frontend/ workspace) named my-project. See catalyst new.

2. Set the database connection

Open backend/.env and point the DATABASE_* block at a SQL database you can reach:

DATABASE_DIALECT = postgres
DATABASE_HOST = localhost
DATABASE_PORT = 5432
DATABASE_USER = postgres
DATABASE_PASSWORD = postgres
DATABASE_SCHEMA = database-schema

The scaffold defaults to PostgreSQL. Each Catalyst app needs its own database or schema — if you run more than one app, give each its own so they don’t clash.

3. Let Catalyst create the tables

In the same backend/.env, turn schema synchronization on:

DATABASE_SYNCHRONIZE = true

With synchronize on, the ORM builds the schema from the entities at boot, so you don’t have to write migrations just to get started.

Caution

DATABASE_SYNCHRONIZE = true is for local development only. It can alter or drop columns to match the entities. Turn it off — and use migrations — before pointing the app at any shared or production database.

Next

Your project is scaffolded and wired to a database. Next, add IAM + OAuth to turn it into an identity provider.