ClawHub
Back to skill

Security audit

Data Analysis

Security checks across malware telemetry and agentic risk

Overview

This is a simple spreadsheet analysis instruction skill with no executable code, persistence, credential access, or hidden behavior found.

Install this if you want a general CSV/Excel analysis helper. Before using it on sensitive datasets, confirm which files the agent will read and where charts, summaries, or transformed outputs will be saved.

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 (1)

Vague Triggers

Medium
Confidence
85% confidence
Finding
The skill description is overly broad and can match many ordinary requests involving files, spreadsheets, statistics, charts, cleaning, or summaries. In an agent system, this increases the chance of inappropriate auto-invocation or routing, causing the skill to engage on requests beyond the user's intent and unnecessarily process potentially sensitive uploaded data.

VirusTotal

65/65 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.