ClawHub
Back to skill

Security audit

opportunities

Security checks across malware telemetry and agentic risk

Overview

This skill coherently generates patent-opportunity reports using PatSnap MCP data and disclosed local files, with some third-party web assets users should be aware of.

Install only if you are comfortable authorizing PatSnap MCP access for patent searches and having the skill create a multi-file local report. Open generated HTML reports knowing they fetch fonts and ECharts from external CDNs unless you modify the templates to use local assets. Use a dedicated report folder for each run.

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 (3)

Context-Inappropriate Capability

Medium
Confidence
97% confidence
Finding
The template loads Google Fonts and jsDelivr-hosted ECharts at render time, which introduces third-party network dependencies into a skill described as MCP/file-generation-only. This can leak report access metadata to external services, break offline/reproducible use, and expose consumers to supply-chain risk if the CDN resource changes or is blocked.

Vague Triggers

Medium
Confidence
84% confidence
Finding
The trigger rule is broad enough to activate on many generic technology-related requests, increasing the chance the skill runs unexpectedly and initiates patent searches, analysis, and file generation without clear user intent. In this skill’s context, that can cause unintended tool use, unnecessary data processing, and unwanted creation of multiple local artifacts.

Missing User Warnings

Medium
Confidence
91% confidence
Finding
The skill is explicitly designed to generate ten output files, including multiple HTML, JSON, CSV, and Markdown artifacts, but it does not prominently warn the user up front that local files will be created. This can surprise users, clutter workspaces, overwrite expected outputs, or expose sensitive query content in saved files, especially because the workflow mandates extensive file generation.

VirusTotal

64/64 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.