ClawHub
Back to skill

Security audit

Devtest Optimizer

Security checks for vulnerabilities and agentic risk

Overview

This skill analyzes user-provided Azure cost and VM data and produces recommendations/templates without directly accessing Azure or running cloud changes.

Before installing, be prepared to share Azure inventory, cost, and subscription exports that may contain sensitive operational details. Do not paste credentials, access keys, or secret values, and review any generated runbook or policy before applying it to Azure.

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
  • MCP Tool PoisoningHidden Instructions, Unicode Deception, Parameter Description Injection
  • 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)

Intent-Code Divergence

Medium
Category
Not specified by scanner
Confidence
81% confidence
Finding

Line L17 explicitly frames the skill as analysis-only with no direct operational behavior. However, the documented workflow and outputs later include generating Azure Automation runbooks and Azure Policy artifacts (L56, L62-L63), which are intended to drive resource state changes and governance actions; this contradicts the claimed purely non-executing, analysis-only framing.

Content

No source excerpt is available for this finding.

Static analysis

No suspicious patterns detected.