ClawHub
Back to skill

Security audit

ServiceNow

Security checks across malware telemetry and agentic risk

Overview

This ServiceNow skill mostly matches its purpose, but its Quick Start includes unrelated SendGrid write commands that could modify the wrong connected account.

Review before installing. The ServiceNow integration itself is disclosed, but do not run the Quick Start commands as written; they reference SendGrid and may alter a SendGrid account if that integration is connected. Prefer read-only ServiceNow verification commands first, and require explicit confirmation before any create, update, delete, checkout, attachment, or CMDB operation.

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 (2)

Intent-Code Divergence

Medium
Confidence
97% confidence
Finding
The Quick Start section for a ServiceNow skill contains SendGrid tool invocations that create or modify external email resources, which directly contradicts the skill’s stated purpose. This can mislead an operator or agent into executing unintended actions against the wrong integration, causing unauthorized writes, data leakage, or operational mistakes in another connected SaaS account.

Missing User Warnings

Medium
Confidence
95% confidence
Finding
The unrelated Quick Start examples perform external write operations against SendGrid without any warning, preview, or confirmation step, despite the rest of the skill emphasizing confirmation for writes. In context, this is dangerous because users may treat Quick Start commands as safe initialization steps and accidentally create contacts or templates in an unrelated service.

VirusTotal

63/63 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.