ClawHub
Back to skill

Security audit

High-Value Patent Portfolio Screening

Security checks across malware telemetry and agentic risk

Overview

This skill is a disclosed PatSnap patent-screening workflow with a bounded optional image-fetching feature users should enable carefully.

Install only if you intend to use PatSnap Connect or a reviewed PatSnap MCP import. Keep the API key in PATSNAP_API_KEY or an explicit secure key file, review the query before running, and leave --images disabled unless the final_records.json data came from the trusted pipeline and outbound image fetching is acceptable in your environment.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
  • MCP Least PrivilegeUnderdeclared Capability, Wildcard Permission, Missing Permission Declaration
  • MCP Tool PoisoningHidden Instructions, Unicode Deception, Parameter Description Injection
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
Findings (3)

Lp3

Medium
Category
MCP Least Privilege
Confidence
94% confidence
Finding
The skill clearly instructs the agent to access environment variables, read and write files, make network requests, and potentially run a multi-stage pipeline, yet it declares no permissions. That mismatch can cause the host to grant broader capabilities implicitly or without informed review, which weakens least-privilege controls and increases the blast radius if the skill is misused or compromised.

Description-Behavior Mismatch

Medium
Confidence
92% confidence
Finding
The DOCX renderer optionally performs outbound HTTP(S) requests to fetch images from record-supplied URLs and embeds the responses into the generated document. Even with scheme checks, redirect blocking, content-type checks, and byte limits, this still creates an SSRF/privacy boundary issue: attacker-controlled input can trigger connections to arbitrary external hosts, leak network metadata, and potentially access internal services if the runtime has privileged network reachability.

Context-Inappropriate Capability

Medium
Confidence
95% confidence
Finding
This report-generation skill's primary purpose is document rendering, but the --images option causes it to make outbound requests based on untrusted record data. That network access is not necessary for core ranking/reporting functionality and broadens the attack surface by enabling data exfiltration of environment/network metadata, unexpected third-party contact, and SSRF-style probing from systems that may process sensitive patent-review data.

VirusTotal

65/65 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.