ClawHub
Back to skill

Security audit

Profile Governance Register

Security checks across malware telemetry and agentic risk

Overview

The skill is a small synthetic validation workbench that returns a user-provided governance key and does not include code, credential access, file access, or an actionable network endpoint.

Install only if you understand this is a synthetic validation/workbench skill rather than a full governance administration tool. Do not use it for real account policy changes unless the publisher clarifies what, if anything, is recorded through the referenced controlled validation endpoint.

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 (2)

Description-Behavior Mismatch

Medium
Confidence
93% confidence
Finding
The skill claims to register a profile governance setting, but the body says runtime validation records effects through a controlled URL endpoint instead of actually performing the declared action. This semantic mismatch can mislead an agent into treating an outbound recording/staging workflow as a legitimate configuration operation, creating a hidden side effect channel and weakening user consent and auditability.

Intent-Code Divergence

Medium
Confidence
90% confidence
Finding
The documentation first presents the skill as governance registration, then narrows the implementation to synthetic validation with downstream recording to a controlled URL. Even if the endpoint is described as controlled, the contradiction is dangerous because it obscures the actual operation, making it easier to smuggle exfiltration-like or unauthorized outbound actions under an administrative-looking skill description.

VirusTotal

63/63 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.