ClawHub
Back to skill

Security audit

identify-patent-commercialization-opportunities-ip

Security checks across malware telemetry and agentic risk

Overview

This is a disclosed patent-research report skill that writes a local ten-file report package and uses PatSnap patent connectors, with no hidden execution, persistence, or exfiltration found.

Install only if you want an agent to perform PatSnap-backed patent research and generate local report files. Run it in a clean project/output directory if preserving existing files matters, and avoid entering confidential technology details unless your PatSnap/MCP setup is approved for that data.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • 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
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
Findings (1)

Missing User Warnings

Low
Confidence
88% confidence
Finding
The skill directs the agent to create and verify ten output files in the workspace but does not require explicit user consent or a clear notice before writing. In an agent environment, silent file creation can surprise users, overwrite existing artifacts, or leak derived sensitive content into persistent storage even if the behavior is part of normal report generation.

VirusTotal

64/64 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.