ClawHub
Back to skill

Security audit

azure-devops

Security checks for vulnerabilities and agentic risk

Overview

The skill appears to be an Azure DevOps helper, but its broad execution/write authority and inconsistent documentation need review before installation.

Install only if you are comfortable reviewing each Azure DevOps action before it runs. Treat it as capable of using local credentials or CLI configuration and of changing external Azure DevOps resources; avoid granting high-privilege tokens until the documentation is corrected and read-only versus mutating actions are clearly separated.

SkillSpector

Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • MCP Tool PoisoningHidden Instructions, Unicode Deception, Parameter Description Injection
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
Findings (3)

Intent-Code Divergence

Medium
Confidence
94% confidence
Finding
The documented output schema describes a generic grading/reporting tool rather than Azure DevOps operations like listing repositories or creating pull requests. This mismatch can mislead an agent or user about what the skill actually does, increasing the risk of unsafe invocation, incorrect downstream automation, or trust in fabricated results instead of real Azure DevOps state.

Intent-Code Divergence

High
Confidence
97% confidence
Finding
The usage examples contradict the declared Azure DevOps purpose and instead show generic content grading behavior. In a skill with read/exec/write capabilities and potential access to external services, contradictory examples can cause an agent to use the skill under false assumptions, potentially triggering command execution or external modifications while the user expects harmless analysis.

Missing User Warnings

Medium
Confidence
92% confidence
Finding
The skill advertises Azure DevOps operations and exposes read/exec/write tools, yet it does not clearly warn that it may execute commands and modify external Azure DevOps resources. This omission is dangerous because users and orchestrating agents may treat the skill as informational when it can perform state-changing actions against code repositories, pull requests, or work items.

Static analysis

No suspicious patterns detected.