ClawHub
Back to skill

Security audit

数据分析师

Security checks across malware telemetry and agentic risk

Overview

This skill is a data-analysis helper with broad local tooling, but the inspected artifact does not show hidden, destructive, or deceptive behavior.

Before installing, treat it as a local data-analysis skill that may read, write, and execute code in the working environment. Use it with datasets and directories you intend to analyze, and review any generated Python or command execution before running it on sensitive files.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
  • 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
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
Findings (2)

Intent-Code Divergence

Medium
Confidence
89% confidence
Finding
The skill is presented primarily as a data-analysis assistant, but later sections describe generic command execution and troubleshooting for arbitrary command failures. This capability expansion can mislead users and agents into invoking shell execution outside the justified scope, increasing the chance of unsafe local command execution or abuse of the enabled exec tool.

Context-Inappropriate Capability

Medium
Confidence
92% confidence
Finding
The skill declares the exec capability even though its stated purpose is data cleaning, statistics, visualization, and report generation, which do not inherently require unrestricted command execution. Granting exec without clear necessity violates least privilege and can enable file-system changes, external process launch, or execution of attacker-influenced commands if the skill is later driven by untrusted input.

VirusTotal

61/61 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.