ClawHub
Back to skill

Security audit

Review Japanese Patent Claims Ip

Security checks across malware telemetry and agentic risk

Overview

This skill is a disclosed Japanese patent-claim review workflow that creates a user-approved static HTML report and does not show hidden, destructive, or unrelated behavior.

Install if you are comfortable providing patent materials to the agent and, when used, to the configured patent-search services. Choose the output directory intentionally and have Japanese patent counsel review any filing, amendment, deadline, translation, or patentability conclusions.

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
92% confidence
Finding
The skill explicitly instructs the agent to create a file in a user-approved output directory, but it does not require a separate user-facing confirmation immediately before writing. That creates a real, though low-severity, file-write risk because an agent could persist content to disk as a side effect in contexts where the user did not fully understand that a file would be created.

VirusTotal

64/64 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.