---
id: ITC-WP-0010
type: workplan
title: "CARING Kubernetes RBAC Benchmark"
domain: canon
repo: info-tech-canon
status: proposed
priority: medium
created: "2026-05-23"
updated: "2026-05-23"
depends_on_workplans:
  - ITC-WP-0003
  - ITC-WP-0005
state_hub_workstream_id: "b64f0fc9-8668-4c02-8247-67a41660bdeb"
---

# ITC-WP-0010 - CARING Kubernetes RBAC Benchmark

## Goal

Create a distinct benchmark workplan for analyzing Kubernetes RBAC through
CARING and the wider InfoTechCanon kernel.

## Intent

This is deliberately separate from the small SaaS proof. The benchmark is more
ambitious and should stress orthogonality across Access Control, Organization,
Governance, Security, Network, DevSecOps, Observability, Task, and Tagging.

## Tasks

### T01 - Benchmark workspace

```task
id: ITC-WP-0010-T01
status: todo
priority: high
state_hub_task_id: "9ad31e13-7dc2-469c-b539-d3375a16c5f4"
```

- Create `infospace/standards/caring/benchmarks/kubernetes-rbac/`.
- Define source corpus, cases, expected outputs, and review criteria.

### T02 - RBAC assimilation

```task
id: ITC-WP-0010-T02
status: todo
priority: high
state_hub_task_id: "180d7ccf-7daa-4f4c-a92a-641ef5d7b442"
```

- Map Kubernetes Role, ClusterRole, RoleBinding, ClusterRoleBinding,
  ServiceAccount, Namespace, verbs, resources, and scopes.
- Preserve the warning that Namespace is not automatically a tenant boundary.

### T03 - CARING access descriptors

```task
id: ITC-WP-0010-T03
status: todo
priority: high
state_hub_task_id: "4ffd6643-a7ab-487c-a09a-0fcaf0115c83"
```

- Express benchmark cases as CARING access descriptors.
- Distinguish declared access, effective access, derived capability, and
  induced access.

### T04 - Findings and canon pressure

```task
id: ITC-WP-0010-T04
status: todo
priority: medium
state_hub_task_id: "52632a4c-6e03-4212-ad6b-0cbb7b3a6e42"
```

- Produce gaps, conflicts, mappings, and proposed canon changes.
- Feed stable findings back into models and standards through explicit tasks.

## Acceptance

- Kubernetes RBAC is analyzed as a benchmark, not as a shortcut profile.
- CARING descriptor shape is tested with practical examples.
- Benchmark findings produce explicit canon change proposals.