Back to skill

Security audit

Find Skills Clark

Security checks across malware telemetry and agentic risk

Overview

This is an instruction-only skill for finding and installing other skills, with no hidden code or data access, but users should confirm the registry and exact install before proceeding.

Before installing, check whether the result comes from skillhub or clawhub, review the exact skill name/version and any risk signals, and explicitly approve the install command only if you trust that source.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
Findings (2)

Natural-Language Policy Violations

Medium
Confidence
94% confidence
Finding
The skill hard-codes routing based on users being Chinese or on CN networks, and directs them to different repositories without asking for consent or explaining the policy. This creates geography/language-based behavioral differences that can affect privacy expectations, autonomy, and trust, especially when the repositories may differ in content, governance, or security posture.

Missing User Warnings

Medium
Confidence
96% confidence
Finding
The skill explicitly says it can install skills for the user, but only requires summarizing source, version, and risk signals before installation; it does not require explicit consent, warn that installation modifies the environment, or require confirmation before executing a system-changing command. In a high-priority discovery flow, this increases the chance of unintentional package installation from external sources.

VirusTotal

64/64 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.