6. Import

Bulk-load assets and their metadata into the catalog in a single operation — from a live database connection or by uploading a file.

The Import section (Data Management → Import) is the primary way to populate your catalog at scale. When you open it, Metaustral presents two paths:

The Import landing — choose between "From a connection" (live database discovery) or "From a file" (Excel template or dbt project).

6.1 From a connection

This path connects to a registered data source and automatically discovers its tables, views, columns, and data types. Before using it you must first create a connection in Data Management → Integrations (see section 5).

The flow has four steps: Connect → Browse schemas → Review → Done.

Choose a connection

All saved connections are listed grouped by type (Databases, BI Tools, File sources). Use the search box to filter by name and click the connection you want to import from.

Browse schemas and select tables

Metaustral connects to the source and lists all available schemas in the left panel. Select a schema to load its tables and views on the right. Assets already in the catalog are marked Already in catalog; new ones are marked New. Select the rows you want to import — you can select all or individual items.

Review and confirm

The review screen shows a summary of new assets to create and existing ones to update, with their name, type, schema, and column count. You can optionally set a default owner and default domain to apply to all newly created assets. Click Confirm import to execute.

Import from connection — step 1: choose a connection

Step 1 — Choose a connection: all registered connections are shown grouped by type. Click any card to proceed.

Import from connection — step 2: browse schemas and tables

Step 2 — Browse schemas: select a schema in the left panel to load its tables and views. Rows already in the catalog are flagged so you can skip or re-import them.

Import from connection — step 3: review and confirm

Step 3 — Review: stat cards show how many assets will be created, updated, or left unchanged. The table lists each asset with its type, schema, and column count before you confirm.

6.2 From a file

This path does not require a live connection. Instead, you upload a structured file that Metaustral parses to create or update catalog assets. Two file formats are supported:

File type selection — choose Excel (.xlsx) for structured bulk import using the Metaustral template, or dbt project to import models, columns, and lineage from your dbt manifest.

6.2.1 Excel (.xlsx)

Download the official Metaustral Excel template directly from the import screen. The template contains multiple sheets — Assets, Columns, Lineage, and ERD — where you fill in your data. Once ready, drag and drop the completed file into the upload zone (or click to browse). The import follows three steps: Upload → Review → Done.

Step 1 — Upload: download the template, fill it in, then drag & drop your .xlsx file into the upload zone.

After uploading, Metaustral validates the file and shows a full preview before writing anything to the catalog. The review screen displays stat cards (new assets, updates, columns, lineage rows, ERD rows, errors) and two tables — assets to create and assets to update — with the exact fields that will change. Any row-level errors must be fixed in the file before the import can run.

Step 2 — Review: stat cards summarize the operation at a glance. The "Assets to create" and "Assets to update" tables show each affected row with name, type, status, and the fields that will be written.

6.2.2 dbt project

If your team uses dbt, you can import models, column documentation, descriptions, and lineage directly from the artifacts generated by your dbt project. Upload two files:

manifest.json (required) — generated by dbt compile or dbt run. Contains models, sources, column docs, and lineage (DAG).
catalog.json (optional) — generated by dbt docs generate. Enriches the import with real column types queried directly from your database.

Step 1 — Upload files: drop manifest.json (required) and catalog.json (optional) into their respective zones, then click "Preview import".

After uploading, a review screen shows stat cards (new assets, updates, columns, lineage rows, errors) and the full list of assets that will be created. Click Execute import to apply the changes to the catalog.

Step 2 — Review: stat cards show the totals. The "Assets to create" table lists each dbt model with its name, type, schema, and description pulled from the manifest.

Tip: imports are always additive or updating — they never delete assets already in your catalog. If a table disappears from the source, the catalog asset is left intact so you don't lose any manually added documentation.