ClawHub
Back to skill

Security audit

Progressive Delivery with Argo Rollouts & Flagger

Security checks for vulnerabilities and agentic risk

Overview

This skill provides coherent Kubernetes deployment workflows, with disclosed operations that can change running applications.

Restrict access to the intended cluster and namespaces, invoke this skill for explicit rollout requests, and establish whether automatic aborts are allowed. Use full promotion deliberately: it skips analysis, pauses, and remaining steps, as documented in the [Argo Rollouts reference](https://argoproj.github.io/argo-rollouts/generated/kubectl-argo-rollouts/kubectl-argo-rollouts_promote/).

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)

Activation keywords include everyday terms such as "promote" and "abort" without requiring Kubernetes rollout context (L020–L023).

Medium
Category
Trigger Abuse
Confidence
93% confidence
Finding

The "When to Apply" section presents "promote" and "abort" as activation keywords without explicitly requiring a Kubernetes deployment request. These words also occur in unrelated requests, so keyword-based activation could invoke this infrastructure-management skill unintentionally.

Content

No source excerpt is available for this finding.

Static analysis

No suspicious patterns detected.