SqlDBM + GitHub

Version your data model the way you version your code. DDL and dbt YAML push straight to your GitHub repository, ready for review.

Book a Demo

THE PROBLEM

Schema changes are the one thing that skips code review

Application code has a path everyone trusts: a branch, a pull request, a reviewer, a merge. Database changes rarely get the same treatment — a DDL script gets pasted into a ticket, or a Slack thread, or run straight against the warehouse by whoever holds the credentials. The change ships, and the repository never learns it happened.

Nobody reviewed it. The person who wrote the DDL is usually the only one who read it.
Nothing to search later. “When did this column change, and why” has no answer in the repo.
Two sources of truth. The model and the deployed schema drift, and nobody’s sure which is right.

THE PATH

Schema takes the route everything else already takes

Generated DDL and dbt YAML leave SqlDBM as a commit, open a pull request against your main branch, and pass through whatever review and checks you already run before anything reaches the database. The same button sits on the Forward Engineering and DataOps screens, so a first build and an incremental alter follow the identical path.

Without the integration, DDL runs by hand straight from the data model to the database, leaving no record in the repository. With SqlDBM and GitHub, the same change passes through the repository and a pull request, then your pipeline, before reaching the database.

WHAT REVIEWERS SEE

A diff, not a diagram

The pull request holds the generated SQL, line by line. Anyone with access to the repository can read exactly which columns are being added, dropped or retyped — no SqlDBM seat needed, and no credentials to the warehouse. The people who review your application code can review the schema it depends on.

A GitHub pull request generated by SqlDBM showing an ALTER TABLE diff adding a PRODUCT_DESCRIPTION column, with 2 approvals required and branch protection active

Connected in four steps

1

Create a personal access token

In GitHub, under Developer settings, generate a token with read and write access to the repository. Copy it when it appears — GitHub won’t show it again.

2

Add the connection in SqlDBM

On the User Connections page, choose GitHub and paste the token.

3

Point it at a repository

Paste the repository’s HTTPS URL. A new repository needs at least one file on its main branch, so initialise it with a README first. Choose whether pushes open a pull request or commit directly.

4

Push

Generate scripts on Forward Engineering or DataOps and select Push to Git. Commits land on the branch you chose; pull requests appear under Pull requests.

THE PAYOFF