Getting started

Import a schema

A read-only script runs in your own environment, produces a JSON description of your schema, and you submit that JSON to ZestreaDB. Nothing connects to your database from our side, and no credentials ever leave your network.

It reads names, types, and dependencies — the shape of your schema, not what is inside it. It never reads table rows, credentials, or procedure bodies. Review the script before you run it: read it yourself, or paste it into your own AI assistant with the prompt below.

Start small. The filter block at the top of the script (@OnlyObjects, @OnlySchemas) lets a first import be the ten or fifteen objects of one flow rather than the whole database. One documented flow is more useful than a catalog nobody can hold in their head, and it keeps the payload small.

For the automated paths, you can wrap the script as a stored procedure in your own database, so the scheduled call is one line instead of piping a long script. You do not need this for a manual import — paste the script and go. We do not ship an installer: if you want the procedure, you wrap it yourself, with your own naming and change control.

The demo workspace documents this flow — load the demo and open the ZestreaDB extraction flow to see what it looks like.