kubara catalogs

This repository contains the catalog sources for kubara.

In kubara, a catalog is the reusable platform input that defines service metadata, Helm charts, Terraform modules, overlays, and other assets used to generate a platform setup.

What is in this repository?

Today this repo contains the official catalogs that kubara uses as its default platform baseline:

• bootstrap/ – the fixed bootstrap foundation, like Argo CD and bootstrap CRDs, used during the initial platform bring-up
• general/ – the general reusable platform services that are meant to be selected and configured per cluster

These catalogs are versioned and meant to be packaged and distributed as OCI artifacts.

Publishing

Catalogs are published independently from git tags. Before creating a tag, update spec.version in the corresponding Catalog.yaml and merge that change to main.

• bootstrap-v1.0.1 publishes bootstrap/ as version 1.0.1
• general-v1.0.1 publishes general/ as version 1.0.1

The tag version must match the catalog manifest exactly. Published versions are immutable, so use a new version instead of moving or recreating a release tag.

Compatibility checks

Catalog changes are packaged locally and exercised through the full kubara init and generate flow with every service enabled before their generated Helm and Terraform output is validated. The compatibility matrix covers every supported provider and Kubernetes type combination: STACKIT SKE, STACKIT Edge, and T Cloud Public CCE. The default kubara revision is pinned in the catalog checks workflow to keep pull request results reproducible.

Maintainers can run the catalog checks workflow manually with a kubara_ref override to test a tag, commit SHA, branch, or pull request ref. The pinned revision should move to the latest compatible stable release once the external catalog model is available in a kubara release.

What is this repo for?

This repo is the maintainable source for the official kubara catalogs.

That means:

• catalog manifests live here
• service definitions live here
• reusable platform stacks live here

kubara can consume these catalogs during schema generation, config initialization, platform generation, and bootstrap flows.

If you want to understand how catalogs fit into the bigger platform model, these docs are the best entry points:

• Catalogs concept
• Catalog templating
• Components overview
• Architecture overview

Community and future direction

We want this repository to be a place for community engagement around reusable kubara platform stacks.

The current catalogs are only the starting point. We are open to growing this into a broader ecosystem with additional catalogs for different needs, for example:

• different infrastructure providers
• different opinionated platform stacks
• lighter or more enterprise-focused setups
• security, observability or developer-platform variants

That can include official catalogs, experimental catalogs, and community-driven catalogs over time.

If you want to propose a new stack, extend an existing catalog, improve service definitions, or contribute a new catalog direction entirely, contributions and discussion are welcome.

Working with these catalogs

If you want to build your own catalog or adapt ideas from this repo:

1. Read How to create a catalog
2. Read Catalog distribution
3. Use this repo as a real example of catalog structure and layout

License

See LICENSE.