ClawHub
Back to skill

Security audit

A professional skill for reading electronic datasheets

Security checks across malware telemetry and agentic risk

Overview

This is a coherent local datasheet/PDF helper with ordinary cautions around reading user-provided PDFs, rendering pages to files, and installing Python PDF libraries.

Install the Python PDF dependencies from trusted sources, preferably in an isolated environment. Use this skill only with PDFs you are comfortable having the agent inspect, and avoid giving render_page sensitive or system-level output paths because it can write the rendered PNG to the path supplied.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
Findings (2)

Context-Inappropriate Capability

Medium
Confidence
88% confidence
Finding
The function accepts an arbitrary out_path and writes rendered image files there without constraining the destination to a safe workspace. If an untrusted caller can influence that path, the tool can overwrite or create files outside the intended directory, enabling data clobbering, persistence, or placement of files in sensitive locations available to the running user.

Vague Triggers

Medium
Confidence
78% confidence
Finding
The trigger text is very broad: it activates on general PDF-plus-chip questions and claims coverage for all IC types, which can cause the skill to engage outside narrowly intended scenarios. Overbroad activation increases the chance the agent processes unintended documents or answers from the wrong context, which can expose sensitive PDF content or produce unsafe, overconfident technical guidance.

VirusTotal

64/64 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.