Security audit

Defender Posture Reviewer

Security checks for vulnerabilities and agentic risk

Overview

This is a read-only guidance skill for analyzing user-provided Microsoft Defender data and producing remediation advice, with no artifact evidence of hidden execution or data theft.

Before using any generated Azure CLI fixes, review them with the responsible Azure owner and apply them in a controlled change process. Provide exports or read-only CLI output, not credentials, access keys, or secret values.

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: 91% confidence
Finding: Line L17 states the skill is instruction-only and does not execute Azure CLI commands or access the Azure account directly. However, the documented workflow and output explicitly include generating Azure CLI commands for remediation (L51, L64), which contradicts the strict 'instruction-only' framing because the skill is still directing command-based changes rather than limiting itself to passive analysis.

Static analysis

No suspicious patterns detected.