ClawHub
Back to skill

Security audit

Gee Dataset Intelligence Skill

Security checks across malware telemetry and agentic risk

Overview

This skill is a coherent Google Earth Engine dataset catalog helper with offline querying by default and clearly disclosed optional refresh and audit operations.

Installers should understand that normal catalog searches are local, but refreshing or auditing the catalog can fetch Google pages, write under assets/ and reports/, and, with explicit LLM flags, send public catalog metadata to a configured third-party API using an environment-provided key.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • MCP Least PrivilegeUnderdeclared Capability, Wildcard Permission, Missing Permission Declaration
  • 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
Findings (1)

Lp3

Medium
Category
MCP Least Privilege
Confidence
90% confidence
Finding
The skill documentation instructs use of environment variables, local file reads/writes under assets/ and reports/, and explicit network operations, but it declares no permissions. This creates a security and governance gap: a host may allow execution without presenting the user with accurate capability warnings, increasing the chance of unintended network access or filesystem modification.

VirusTotal

65/65 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.