Resources

Rise carries a small generic resource substrate that backs Organizations, ResourceDefinitions, and (eventually) any custom kind an external controller wants to reconcile. The model is intentionally Kubernetes-shaped — typed objects with apiVersion / kind / metadata / spec / status, controller-owned status keys, finalizer-gated deletion, optimistic concurrency on revision — without making Kubernetes itself a dependency.

The contract was introduced together with multi-organization support: an Organization is just a built-in root-scoped resource, and the same substrate that stores it can store anything else a controller registers.

When to use it

• You’re writing a controller that reconciles state outside Kubernetes (Snowflake, ECR, AWS RDS, internal services). Register a ResourceDefinition, persist spec, write under status.controllers["your-id"]. See Custom Resources.
• You’re an operator managing built-in objects (Organizations, ResourceDefinitions). Use the HTTP API directly. See HTTP API.
• You need to understand storage semantics — finalizers, cascade deletion, revision conflicts. See Storage Model.

What’s in scope today

Built-in resources	Organization, ResourceDefinition
External custom resources	Yes, via ResourceDefinition
HTTP API access tiers	Operator (full CRUD), Controller (own status/finalizers)
Status writes by end users	No (operator-only API in v1)
Watch / change feed	Not yet — controllers poll list
Pagination / selectors	Not yet — list returns everything
Version conversion	Storage-version writes only; reads/lists project across served versions

End-user / Organization-scoped RBAC is deferred. The generic API is operator-only in v1; admins on the default Organization do not implicitly receive Operator. Controllers may use a dedicated controller-JWT auth context for status/finalizer writes, but cannot perform operator CRUD.

Section guide

• Storage Model — the resources table, finalizers, cascade deletion, garbage collection, revision/concurrency semantics, organization namespace-prefix behavior.
• HTTP API — path grammar, auth tiers, request/response envelopes, status and finalizer subresources.
• Custom Resources — registering a ResourceDefinition, schema validation, version lifecycle, controller migration responsibilities.
• Schema Reference — generated JSON Schemas for Resource, ResourceMetadata, Organization, ResourceDefinition, and the controller status map.