Register a new Data Product
Data Product Initialization and Registration
Initializing a data product involves connecting it to a Git repository and publishing it for the first time. You can either create a new repository directly from Blindata or connect an existing repository from your Git provider. Each data product must have a descriptor file (DPDS) stored in the repository; the descriptor becomes the authoritative definition of the product.
This page describes the workflow from the catalog into the Builder. The only other path is from a Blueprint , which has dedicated documentation.
Initialize from the catalog
The initialization flow starts from the Data Product Catalog . You create or select a data product there, then open its Builder to configure the repository and descriptor.
1. Create or select the data product
Navigate to the Data Product List page:
Click the button in the bottom-right corner to create a new data product:
Provide the general information (name, identifier, description, metadata). For an existing product, open it from the list.
2. Open the Builder
From the data product detail page, click BUILDER to open the Builder and connect the data product to Git:
3. Configure repository in the Builder
When no repository is configured yet, the Builder shows a warning banner. Click CONFIGURE REPOSITORY to start:
Before or during the configuration, you must provide valid Git provider credentials via a personal secret. Then choose how to link the repository and, if needed, adjust advanced settings.
Personal secret for the Git provider
The Builder uses your personal secret (e.g. a GitHub Personal Access Token) to access the Git provider.
- Open the Platform dropdown in the top-right corner of the Builder.
- Select Manage your secrets.
- In the Secrets configuration modal, create or edit a secret (for example,
github-pat):
- Secret Name: e.g.
github-pat - Expiration Date: Optional.
- Secret: The token value (GitHub PAT or equivalent).
- Description: Optional notes.
All selected secrets must be valid before proceeding; expired secrets will block the configuration.
Repository configuration
In the Repository Configuration modal, choose how to link the data product to Git:
- Create New Repository — Blindata creates a new repository and initializes it with the descriptor.
- Select Existing Repository — Use an existing repository and specify where the descriptor is located.
Select the secret you created (e.g. github-pat) in the Configure secrets section.
For Create New Repository:
- Organization — Select the organization or namespace.
- Repository Name — e.g. the fully qualified name of the data product.
- Repository Description — Optional.
- Repository Visibility — Private or public.
Advanced configuration
Expand Advanced configuration to control how the descriptor is stored and whether initialization is skipped:
| Setting | Description |
|---|---|
| Descriptor Branch | The branch where the descriptor file lives. Default is main. |
| Descriptor Path | The file path to the descriptor (e.g. src/main/descriptor.json). Use a custom path when the repository hosts more than one data product — each product can point to a different descriptor file. |
| Skip repository initialization | When checked, Blindata does not initialize the repository. Use this when the repository already exists and you only need to register it. |
All fields have defaults; change them only when you need multiple products per repository or a custom layout.
Initialize from a Blueprint
To initialize a data product using a blueprint, refer to the Blueprint page for detailed instructions.