Skip to content

docs: remove stale references to ActorTemplate as a Kubernetes CRD - #1404

Open
Youssuf Elshall (yelshall) wants to merge 1 commit into
agent-substrate:mainfrom
yelshall:docs/actortemplate-post-crd-cleanup
Open

docs: remove stale references to ActorTemplate as a Kubernetes CRD#1404
Youssuf Elshall (yelshall) wants to merge 1 commit into
agent-substrate:mainfrom
yelshall:docs/actortemplate-post-crd-cleanup

Conversation

@yelshall

@yelshall Youssuf Elshall (yelshall) commented Sep 2, 2026

Copy link
Copy Markdown
Contributor

What

Now that the ActorTemplate CRD has been deleted and its resources moved to the substrate gRPC API and the control-plane store (created/managed with kubectl-ate, persisted in PostgreSQL), several documents still describe ActorTemplate as a Kubernetes CRD, or describe namespace/RBAC relationships that no longer exist. This sweeps the docs for those stale references.

Fixes #368 (docs side).

This change was prepared with AI assistance; I have reviewed and tested it.

  • Docs and comment-only change; no functional code changed, no tests affected

The ActorTemplate CRD was deleted when its resources moved to the
substrate gRPC API and the control-plane store (created and managed
with kubectl-ate, persisted in PostgreSQL). Several documents still
described ActorTemplate as a Kubernetes CRD, or described
namespace/RBAC relationships that no longer exist:

- architecture.md: the resource-model section, the persona
  description, the governance rationale, and the UML class diagram
  all placed ActorTemplate in kube-apiserver as a CRD. The diagram
  also showed a "workerPoolRef" relation that does not exist; pool
  selection is a label-based workerSelector match, and an Actor
  references its template with the actor_template ref.
- glossary.md: ActorTemplate was listed under "Resources (declarative,
  Kubernetes CRDs)", and the Atespace entry contrasted atespaces with
  "the namespace an ActorTemplate lives in" (templates now live in
  atespaces; WorkerPools live in namespaces).
- api-guide.md: the Best Practices section required ActorTemplate and
  WorkerPool to share a namespace or RBAC; placement is actually by
  sandboxClass and workerSelector labels. The ActorTemplate section
  also kept CRD-era spec.-prefixed field paths ("ActorTemplateSpec",
  "spec.volumes", "spec.resources") even though the protojson
  manifest has top-level fields, and typed workerSelector as
  *LabelSelector where the proto type is Selector.
- csi-volumes.md: same stale spec.-prefixed field paths, which
  contradicted the top-level fields shown in its own YAML examples.
- api-style-guide.md: cited actor_template_name as an immutable field
  example; the field is actor_template, and it is mutable - the
  actor's immutable example is source_snapshot_tag.
- code-layout.md: described cmd/atecontroller as the controller for
  WorkerPool/ActorTemplate; it now reconciles WorkerPools and
  network/egress resources only.
- code comments: template_reconciler.go's goldenSnapshotWarmupFor
  said it "mirrors the CRD controller's function of the same name;
  keep both in sync" - the CRD controller was deleted, so that
  instruction referenced nonexistent code. The claude-code-multiplex
  demo UI comment labeled ActorTemplates as k8s CRDs, and
  benchmarking/workloads/deploy.sh used a spec.-prefixed field path
  in a comment (the protojson manifest is top-level).
- README.md: demo blurbs referenced "CRD routing" and templates
  "across three namespaces"; routing goes through the Substrate
  router, and the multi-template demo places templates in different
  atespaces.
@yelshall
Youssuf Elshall (yelshall) force-pushed the docs/actortemplate-post-crd-cleanup branch from 2d166d5 to 1b34a49 Compare September 2, 2026 18:21
@zoez7

Copy link
Copy Markdown
Collaborator

I see this PR is still in draft state, is this ready for review?

@yelshall

Copy link
Copy Markdown
Contributor Author

I see this PR is still in draft state, is this ready for review?

Yes, ready for review now

@yelshall
Youssuf Elshall (yelshall) marked this pull request as ready for review September 3, 2026 10:17
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

Decouple substrate-plane resources from Kubernetes CRDs

2 participants