ClawHub
Back to skill

Security audit

Multi Unit Test Report

Security checks across malware telemetry and agentic risk

Overview

This is an instruction-only report-writing skill, but it depends on one specific local Obsidian/iCloud note that users should review before use.

Before installing, check whether the referenced Obsidian/iCloud mission-statement file exists and contains only report-format guidance you are comfortable sharing with the agent. Prefer a bundled template or user-provided file if you do not want skills reading from personal notes folders.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
  • 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)

Context-Inappropriate Capability

Medium
Confidence
95% confidence
Finding
The skill requires reading a hard-coded file from a local Obsidian vault before doing its stated job, which expands access beyond what is necessary to generate a report. This creates an unnecessary data exposure path and can cause the agent to ingest unrelated sensitive local content from a user-specific filesystem location.

VirusTotal

63/63 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.