ClawHub
Back to skill

Security audit

EZ Ansible Skill

Security checks across malware telemetry and agentic risk

Overview

This is an instruction-only Ansible helper whose risky parts are expected for infrastructure automation, but users should review generated playbooks before running them.

Install is reasonable if you want Ansible generation help. Treat all generated YAML as infrastructure code: review it, run syntax-check/lint/check mode first, limit the inventory, avoid disabling host-key checks without reason, and never commit plaintext vault values or password files.

SkillSpector

By NVIDIA
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 (2)

Intent-Code Divergence

Medium
Confidence
96% confidence
Finding
The skill explicitly recommends using Ansible Vault for secrets, but its example vault file contains realistic plaintext secret values. Even if presented as a template, this normalizes embedding live-looking credentials in source-controlled examples and increases the risk that users will copy them into real repositories before encryption, causing secret disclosure.

Missing User Warnings

Medium
Confidence
72% confidence
Finding
The skill is designed to generate automation that can change files, services, packages, and privileges on remote systems, yet it lacks an explicit safety warning about operational impact. In this context, omission of guardrails can lead users to apply generated automation to production systems without adequate review, increasing the chance of outages or unsafe privilege use.

VirusTotal

64/64 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.