Security audit

azure-cli

Security checks for vulnerabilities and agentic risk

Overview

This skill appears to be an Azure CLI helper, but its instructions are broad and inconsistent enough that users should review it before letting an agent use cloud credentials or run commands.

Install only if you intend to let an agent help with Azure CLI tasks. Before use, require the agent to show the exact az command, target subscription/resource group, expected changes, and cost impact, and approve state-changing commands manually. Avoid using it for code review or generic workflow tasks because the documentation mixes those concepts with cloud operations.

SkillSpector

Vulnerability Patterns

Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
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
Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access

Findings (5)

Description-Behavior Mismatch

High

Confidence: 95% confidence
Finding: The skill is presented as an Azure CLI management capability, but its documented behavior and use cases describe code review, quality scoring, and workflow assistance instead. This kind of semantic mismatch can cause an agent or user to invoke the skill under false assumptions, leading to unintended command execution or inappropriate trust in outputs for privileged cloud operations.

Description-Behavior Mismatch

High

Confidence: 97% confidence
Finding: The declared result schema returns grading and review artifacts rather than Azure resource or command results, which directly contradicts the skill's claimed operational role. In an agent setting, inconsistent output contracts can mislead orchestration logic, causing unsafe follow-on actions, incorrect validation, or silent misuse of privileged execution capabilities.

Intent-Code Divergence

Medium

Confidence: 83% confidence
Finding: The file states a broad workflow/decision-assistance function while later instructing Azure CLI installation and cloud resource management. Contradictory intent increases the chance that an autonomous agent will over-broaden when the skill is appropriate, bypassing human review for operations that can change cloud state or incur cost.

Vague Triggers

Medium

Confidence: 81% confidence
Finding: The invocation guidance is too vague to define when the skill should or should not be used. For a skill exposing read/exec/write tools and cloud-management semantics, ambiguous triggering can cause accidental activation in unrelated contexts, increasing the risk of unintended command execution or credentialed cloud actions.

Missing User Warnings

Medium

Confidence: 90% confidence
Finding: The skill describes Azure resource management and command execution without prominently warning that it can create, modify, or delete cloud resources and potentially incur charges. In the context of a cloud CLI skill with exec capability, missing safety warnings materially increases the likelihood of destructive or costly actions being taken without informed consent.

Static analysis

No suspicious patterns detected.