Task

SkillCloud & infra

Reviews Kubernetes configurations to prevent data exfiltration by AI agents.

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-agents-data-exfil/SKILL.md and read by ahel’s review.

Review infrastructure for AI agent data exfiltration risks (e.g. via prompt injection).

Checks

1. Strict Egress Allowlisting

  • NetworkPolicies: Require restrictive egress. Flag missing egress controls or 0.0.0.0/0.
  • Allowlisting: Limit egress strictly to necessary internal services and authorized LLM APIs.

2. Egress Gateways

  • Interception: Require outbound traffic routing through transparent proxies/Egress Gateways (e.g., Service Mesh, HTTP_PROXY) for Deep Packet Inspection and SNI allowlisting.

3. Data Access

  • Blanket Permissions: Flag broad get/list/watch RBAC on sensitive resources (secrets, configmaps).
  • Over-privileged Mounts: Flag broad or shared sensitive volume mounts.

Signals

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