Azure Networking Audit

• Outcome: Pass
• Latest audit: Sep 14, 2026 · 10:57 PM UTC
• Version: 0.1.1

Overview

This is a read-only Azure networking audit guide with some broader but disclosed Azure inventory checks users should scope carefully.

Install only for users who intend to inspect Azure networking resources with an authenticated Azure CLI session. Prefer granular read permissions and a specific subscription/resource-group scope; avoid running the RBAC and Azure AD identity commands unless you specifically need to confirm who is signed in or verify role assignments.

A.I.G

Vulnerability Patterns

• Skill Instruction Hijacking: Alters the agent's session goals or safety constraints when the skill loads
• Agent Memory Poisoning: Writes attacker-controlled rules into memory that affect later sessions
• Remote Payload Retrieval and Execution: Fetches external code whose behavior can change after review
• Embedded Malicious Code: Ships malicious scripts inside the skill and executes them locally
• Unauthorized Access and Privilege Escalation: Obtains permissions beyond the task's legitimate needs

SkillSpector

Vulnerability Patterns

• Privilege Escalation: Excessive Permissions, Sudo/Root Execution, Credential Access
• MCP Tool Poisoning: Hidden Instructions, Unicode Deception, Parameter Description Injection
• Prompt Injection: Instruction Override, Hidden Instructions, Exfiltration Commands
• Data Exfiltration: External Transmission, Env Variable Harvesting, File System Enumeration
• Supply Chain: Unpinned Dependencies, External Script Fetching, Obfuscated Code

Findings (6)

Description-Behavior Mismatch

Medium

• Confidence: 93% confidence
• Finding: The introductory scope states that application-layer routing in Application Gateway path rules is out of scope, but the document still includes Application Gateway placement in scope and later directs operators to assess Application Gateway instances and utilization. That creates a semantic mismatch in what the skill claims to cover versus what its own procedure and report template ask it to evaluate.

Intent-Code Divergence

Medium

• Confidence: 91% confidence
• Finding: The documentation says application-layer routing in Application Gateway path rules is out of scope, but later sections instruct the operator to identify underutilized Application Gateway instances and report Application Gateway utilization. This is not merely incomplete documentation; it actively points the user to perform analysis on a component that the scope statement says is excluded.

Description-Behavior Mismatch

Medium

• Confidence: 96% confidence
• Finding: The skill is declared as a VNet networking audit, but this reference adds Azure AD identity lookups and subscription-wide RBAC enumeration. That expands access from network configuration review into user and permission discovery, which can expose sensitive identity metadata and enable unnecessary privilege mapping beyond the stated task scope.

Excessive Permissions

Low

• Category: Privilege Escalation

Content

## Prerequisites

- **Azure CLI** authenticated (`az account show` succeeds)
- **RBAC permissions** — Reader role on target subscription, or granular read permissions: `Microsoft.Network/virtualNetworks/read`, `Microsoft.Network/networkSecurityGroups/read`, `Microsoft.Network/azureFirewalls/read`, `Microsoft.Network/expressRouteCircuits/read`, `Microsoft.Network/virtualNetworkGateways/read`, `Microsoft.Network/routeTables/read`, `Microsoft.Network/networkInterfaces/read`
- **Target scope identified** — specific subscription, resource group(s), and VNet name(s). Multi-subscription audits require `az account set` per subscription
- **Network Watcher enabled** — NSG Flow Logs and effective security rules require Network Watcher in the target region. If disabled, document as Critical

• Confidence: 60% confidence
• Finding: Skill requests more permissions than appear necessary for its stated functionality. Review if elevated access is justified.

Description-Behavior Mismatch

Low

• Confidence: 84% confidence
• Finding: The manifest describes a read-only Azure VNet networking audit focused on VNet design, NSGs, firewall, connectivity, peering, and UDR validation. Step 6 extends the skill into broader cost optimization and Azure Advisor recommendation review, which goes beyond the narrowly described networking-audit scope even though it remains read-only.

Description-Behavior Mismatch

Low

• Confidence: 82% confidence
• Finding: The manifest promises a VNet networking audit centered on topology, security controls, connectivity, and routing validation. The 'Resource Optimization' section adds orphaned resource discovery, public IP inventory, Application Gateway capacity review, and Azure Advisor cost recommendations, which are adjacent operational/cost-review tasks rather than the specific networking audit behaviors described.

Static analysis

No suspicious patterns detected.