ClawHub

kubectl

Security checks across malware telemetry and agentic risk

Overview

This is a transparent kubectl helper skill, but it can affect any Kubernetes cluster your current kubeconfig can access.

Install only if you intend to let the agent assist with Kubernetes administration. Before running commands, verify the active kubectl context and namespace, use least-privilege kubeconfig credentials, prefer dry-run modes, and require explicit approval for production changes, delete, drain, exec, cp, rollout, or config-view operations.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • Behavioral ASTexec() Call, eval() Call, Dynamic Import
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
Findings (3)

Missing User Warnings

Medium
Confidence
78% confidence
Finding
Advertising a node drain helper script without an explicit warning understates the operational risk of draining a node, which can evict pods, trigger rescheduling, and cause service disruption if workloads are not disruption-tolerant. In a skill intended for agent-driven execution, this omission is more dangerous because an automated system may invoke the script without appreciating production impact.

Missing User Warnings

Medium
Confidence
94% confidence
Finding
This section advertises mutating and disruptive kubectl operations such as apply, patch, scale, drain, taints, and labels without prominent warnings about their effect on live clusters. In context, this skill is specifically for cluster administration, so omission of safety guidance makes accidental service disruption, rollout mistakes, or node eviction more likely.

Missing User Warnings

Medium
Confidence
92% confidence
Finding
The documented use of kubectl logs and kubectl exec can expose secrets, tokens, environment variables, application data, and internal system state from running containers. Because the skill presents these actions as routine debugging steps without data-handling warnings, it increases the risk of sensitive information disclosure or unsafe shell access to privileged workloads.

VirusTotal

66/66 vendors flagged this skill as clean.

View on VirusTotal