ClawHub
Back to skill

Security audit

Infrastructure Drift Detector

Security checks for vulnerabilities and agentic risk

Overview

This skill is a coherent infrastructure drift-auditing guide, but users should review cloud account scope and any generated remediation or monitoring commands before running them.

Install only if you want an agent to inspect infrastructure state using your local IaC files and configured cloud credentials. Review any generated CI schedules, cron jobs, terraform import commands, or terraform apply commands before enabling or running them.

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
Confidence
78% confidence
Finding
At L031 the text characterizes the operation as read-only, but elsewhere in the same documented skill the CloudFormation path uses `aws cloudformation detect-stack-drift` (L083), which initiates a service-side operation rather than merely reading existing state. That is an intent-level contradiction in the skill documentation because the file presents the skill as passive detection while including an action that changes remote service state by starting a detection run.

Static analysis

No suspicious patterns detected.