ClawHub
Back to skill

Security audit

Azure Infra Free

Security checks across malware telemetry and agentic risk

Overview

This skill is a disclosed read-only Azure CLI helper, with some sloppy duplicated text and unnecessary API-key boilerplate but no evidence of hidden, destructive, or deceptive behavior.

Install only if you are comfortable with the agent using your local Azure CLI login to run read-only Azure queries. Review the current subscription before use, avoid pasting real API keys into chat or shell examples, and treat the duplicated boilerplate as a documentation quality issue rather than a security feature.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • 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
Findings (3)

Intent-Code Divergence

Medium
Confidence
92% confidence
Finding
The skill states that secrets must not be output or recorded, but later instructs users to configure an API key via an environment variable using a literal example. This creates inconsistent security guidance and increases the chance that users or downstream agents will handle, echo, or persist sensitive credentials in terminal history, logs, screenshots, or copied session transcripts.

Vague Triggers

Medium
Confidence
83% confidence
Finding
The trigger condition activates on broad Azure-related keywords and read-only query phrasing without sufficiently constraining scope. Overbroad activation can cause the skill to run in unrelated contexts, prompting unnecessary local command execution and access to Azure account context when the user did not clearly intend that behavior.

Vague Triggers

Medium
Confidence
80% confidence
Finding
The repeated trigger section remains ambiguous and broad, reinforcing auto-invocation behavior without precise boundaries. Duplicate, loosely scoped triggers increase the likelihood of accidental execution paths and make it harder for implementers to apply consistent safety gating.

VirusTotal

62/62 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.