ClawHub
Back to skill

Security audit

qgis

Security checks across malware telemetry and agentic risk

Overview

This is a coherent QGIS helper for local geospatial processing, with disclosed file-writing behavior and no evidence of hidden persistence, credential use, or exfiltration.

Install this only if you want an agent to help run QGIS CLI commands on local GIS data. Before allowing execution, verify the input files, output paths, CRS, units, format, and overwrite policy, especially for batch jobs.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • 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
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
Findings (2)

Vague Triggers

Medium
Confidence
88% confidence
Finding
The default prompt is broad enough to encourage automatic use of the QGIS skill for any geospatial-related request without clearly constraining when invocation is appropriate or what safety checks must occur first. In a file-processing skill, this increases the chance of overbroad tool execution, unintended handling of local files, or acting on ambiguous user requests without explicit confirmation.

Vague Triggers

Medium
Confidence
94% confidence
Finding
Enabling implicit invocation without documented guardrails allows the agent to trigger this local geospatial processing capability automatically, even when the user's request is underspecified. Because QGIS workflows can read, transform, and overwrite local vector or raster data, unrestricted implicit use raises the risk of unintended file access, data modification, and unsafe execution chains.

VirusTotal

65/65 vendors flagged this skill as clean.

View on VirusTotal