Database

Each project gets a Postgres database. The agent reads your schema and generates queries, migrations, and security policies that match your actual tables — no hallucinated column names.

Creating tables

Describe the table in chat:

Create a posts table with title, content, author_id (references users),
published boolean, and created_at timestamp.

The agent writes the SQL migration and applies it. You can also create tables directly from the schema viewer.

Querying data

The agent generates typed queries for your frontend. Here's what it might produce for a blog feed:

const { data: posts } = await supabase
  .from('posts')
  .select('*, author:profiles(name, avatar_url)')
  .eq('published', true)
  .order('created_at', { ascending: false });

Row-level security

RLS policies control who can read, insert, update, and delete rows. The agent writes them from plain descriptions:

Users can only edit their own posts.
Anyone can read published posts.

Migrations

Every schema change is tracked as a migration. Browse migration history from the editor and roll back if something goes wrong.