ClawHub
Back to skill

Security audit

WorkOS

Security checks across malware telemetry and agentic risk

Overview

The skill is a straightforward WorkOS API helper, but it includes high-impact user deletion instructions without enough safety guidance or explicit disclosure in the top-level description.

Install only if you expect the agent to administer WorkOS with a privileged API key. Require explicit confirmation before any create or delete operation, verify organization and user IDs carefully, and prefer reversible account actions where possible.

SkillSpector

By NVIDIA
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 (2)

Description-Behavior Mismatch

Medium
Confidence
85% confidence
Finding
The skill documents destructive user deletion behavior without disclosing that capability in the manifest description, increasing the chance that an agent or operator invokes it unexpectedly. Because deletion is irreversible or operationally disruptive, under-declaring this action weakens informed consent and safe tool selection.

Missing User Warnings

Medium
Confidence
94% confidence
Finding
Providing a direct DELETE example for user accounts without any warning, confirmation requirement, or recovery guidance makes accidental destructive actions more likely. In an agent skill context, omission of guardrails around destructive operations is dangerous because agents may execute examples literally when asked to manage users.

VirusTotal

65/65 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.