Catalog

Back to index

What The Catalog Does

The catalog is the registry-backed discovery surface for packages.

It combines:

• configured registries from platform settings
• workspace-known suites
• built-in example module metadata

How Discovery Works

For each configured registry, the catalog service:

1. calls GET /v2/_catalog
2. enumerates repositories
3. calls GET /v2/<repo>/tags/list
4. builds a package view from the repository plus any locally known metadata

If a discovered repository matches a known workspace suite, the catalog can enrich the registry result with:

• suite title
• owner
• description
• module tags
• pull and fork commands
• inspectability

Package Fields

A catalog package currently exposes:

• id
• kind
• title
• repository
• owner
• provider
• version
• tags
• description
• modules
• status
• score
• pullCommand
• forkCommand
• inspectable
• starred

Favorites

The catalog also supports favorites:

• GET /api/v1/catalog/favorites
• POST /api/v1/catalog/favorites/{packageId}
• DELETE /api/v1/catalog/favorites/{packageId}

These let the frontend mark commonly used packages without changing the registry source itself.

Frontend Route

The main browser is:

• /catalog

This page is the discovery layer for registry-backed content.

Relationship To Suites

The catalog and the runnable suite inventory are related, but not identical:

• Catalog is what registries advertise
• Suites are what BabelSuite can currently inspect and run locally

That distinction is why a package can appear in the catalog even when it is not the current workspace source of truth for runnable suite details.