ClawHub
Back to skill

Security audit

Cilium & Hubble Network Observability

Security checks for vulnerabilities and agentic risk

Overview

This skill provides coherent Kubernetes networking guidance, with disclosed installation and policy changes that can affect cluster connectivity.

Use this skill for explicit Kubernetes networking requests. Confirm the target cluster, namespace, and workload labels before applying examples. Cilium installation and default-deny or clusterwide policies can disrupt connectivity; review changes, test in staging, and retain a rollback plan. Hubble output may contain sensitive traffic details.

A.I.G

Vulnerability Patterns
  • Skill Instruction HijackingAlters the agent's session goals or safety constraints when the skill loads
  • Agent Memory PoisoningWrites attacker-controlled rules into memory that affect later sessions
  • Remote Payload Retrieval and ExecutionFetches external code whose behavior can change after review
  • Embedded Malicious CodeShips malicious scripts inside the skill and executes them locally
  • Unauthorized Access and Privilege EscalationObtains permissions beyond the task's legitimate needs

SkillSpector

Vulnerability Patterns
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
Findings (1)

Standalone triggers such as "flow" and "connectivity" can activate the skill outside Kubernetes networking contexts.

Low
Category
Trigger Abuse
Confidence
96% confidence
Finding

The activation list explicitly treats mentions of generic terms such as "flow" and "connectivity" as sufficient to invoke the skill. Although the description identifies Kubernetes networking as its purpose, these trigger conditions do not require that context or exclude unrelated uses.

Content

No source excerpt is available for this finding.

Static analysis

No suspicious patterns detected.