Task

SkillCloud & infra

Reviews Kubernetes RBAC configurations for permissions and privilege escalation risks.

Available today. Use it from your connected AI after setup.

Connect ahel once, and every AI you use reads what you have installed.

Then ask your AI: use the Task skill

What this skill tells your AI

The instructions your AI receives, as published by gke-labs/kube-agents in .agents/skills/review-security-k8s-rbac/SKILL.md and read by ahel’s review.

Review Kubernetes RBAC configurations for vulnerabilities, prioritizing GKE best practices and advanced attack vectors.

Checks

1. Global & Default Configs

  • Default Groups/Users: Flag non-default bindings for system:anonymous, system:unauthenticated, system:authenticated, system:masters.
  • Broad Service Accounts: Flag bindings for system:serviceaccounts or system:serviceaccounts:<namespace>.
  • Cluster Admin: Flag cluster-admin bindings to unauthenticated/broad groups.
  • System Roles: Flag delete/update on system: prefixed bindings.
  • Missing Namespaces: Flag ClusterRoleBindings to ServiceAccount missing a namespace.

2. Privilege Escalation & Bypasses

  • RBAC & CSR: Flag bind, escalate, create, update, patch on rbac.authorization.k8s.io. Flag create on certificatesigningrequests or update on certificatesigningrequests/approval.
  • Webhooks & CRDs: Flag create, update, patch on mutatingwebhookconfigurations, validatingwebhookconfigurations, or customresourcedefinitions.
  • Auth Probing: Flag create on tokenreviews/subjectaccessreviews.
  • Policy Bypass: Flag update/patch on namespaces (PSA bypass) or modification access to networkpolicies.

3. Role & Binding Design

  • Wildcards: Flag * in apiGroups, resources, verbs. Flag deletecollection (DoS risk).
  • Escalation Verbs: Flag escalate, bind, impersonate.
  • Over-scoped: Flag ClusterRoleBindings used where namespace-local RoleBindings suffice. Flag disjoint resource/verb sets in single roles.
  • Self-Modification: Flag roles allowing pods to self-modify (e.g., updating own ServiceAccount/rolebindings).

4. Sensitive Resources

  • Highly Sensitive: Flag access to secrets, pods/exec, pods/portforward, pods/attach, nodes/proxy (unless strictly scoped via resourceNames).
  • Secret Harvesting: Flag list or watch on secrets.
  • Ephemeral/Status: Flag update/patch on pods/ephemeralcontainers or /status subresources.

5. GKE RBAC Guide

  • Refer to resources/gke-rbac.md for additional guidance. Identify if any of this guidance is both relevant and missing from your review and include those findings as well.

Signals

GitHub stars
54
Forks
36
Last commit
Sep 2026
Advanced
Catalog kind
skill
Gateway key
review-security-k8s-rbac
Source
github.com/gke-labs/kube-agents