Document the flows that run through your database.

Across tables, procedures, jobs, and the context that lives in someone's head— and it tells you when they change.SQL Server today; Postgres next.

No account, nothing to install

Anyone can document it, and nobody has to learn how. You fill in fields — the description, the gotchas, what to do when it fails. And every flow comes out in the same format. This is why it survives the year, and a second pass then isn't a new project.

What it does

Describe a table once, not in every flow

Describe an object once. It shows up in every flow that uses it — not as copies that drift apart.

When the database changes, the documentation knows

Re-import and anything that vanished is flagged. You find out which part stopped being true, not a year later.

Built from your actual database

Nothing connects to your database. You bring the schema in yourself, then map how the data moves — with real dependencies rather than remembered ones.

The context nobody wrote down

Why a table exists, what a status code actually means, and what to do at 2am when the job fails. Written on the object, not in a page nobody opens.

The flow doesn't stop at the database

Add the API, the nightly job, the vendor feed. The pieces of the flow that aren't in your schema but decide what it means.

A document, not just a diagram

Copy a flow as markdown — purpose, objects, connections, business rules. Readable by a person, parseable by an AI.

Share a flow with someone

A read-only link. No account on the other end.