ClawHub
Back to skill

Security audit

PBE Extractor

Security checks across malware telemetry and agentic risk

Overview

This is an instruction-only text analysis skill with one wording inconsistency about review-file handling, but no automatic code, install hook, network call, or hidden persistence behavior.

Safe for normal text-analysis use. Avoid pasting confidential content unless your agent and model provider are approved for it, and only allow creation of any review file if you explicitly want that local artifact.

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 (1)

Intent-Code Divergence

Medium
Confidence
88% confidence
Finding
The skill explicitly assures users it does not write files to disk, but later documentation references adding drift cases to `requires_review.md`, which implies file output. This inconsistency can mislead operators about side effects and trust boundaries, and in some agent environments could cause unexpected persistence of sensitive content or analysis artifacts.

VirusTotal

64/64 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.