ClawHub
Back to skill

Security audit

仪表盘构建(专业版)

Security checks across malware telemetry and agentic risk

Overview

The skill is a coherent dashboard-building helper, but it asks for shell execution and sensitive data credentials without enough enforceable scoping or user-control safeguards.

Review generated shell/Python scripts before running them, use read-only and narrowly scoped API/database credentials, avoid committing .env files or dashboard outputs containing sensitive data, and confirm webhook destinations before enabling alerts.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • MCP Tool PoisoningHidden Instructions, Unicode Deception, Parameter Description Injection
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
Findings (3)

Intent-Code Divergence

Medium
Confidence
93% confidence
Finding
The skill claims command execution is restricted by a whitelist, but the document elsewhere instructs the agent to create and run local shell scripts, Python commands, and fetch jobs without defining any actual enforcement boundary. This mismatch is dangerous because users or downstream agents may trust the safety claim and execute generated commands that can read local secrets, invoke network calls, or modify files.

Vague Triggers

Medium
Confidence
86% confidence
Finding
The skill’s activation language is broad and generic, covering common analytics and visualization requests without strong constraints on when the skill should or should not engage. In an agent environment with Read/Write/Edit/Bash tools, this can cause the skill to trigger inappropriately and steer the agent into credential handling, script generation, or command execution for loosely related prompts.

Missing User Warnings

Medium
Confidence
91% confidence
Finding
The skill instructs users to export API keys and database connection strings and then run fetch scripts, while also supporting external API access and webhook notifications. Although there are later generic security notes, this execution path lacks an immediate, prominent warning about sensitive credential exposure, outbound data transfer, and the need to review generated scripts before execution.

VirusTotal

VirusTotal findings are pending for this skill version.

View on VirusTotal

Static analysis

No suspicious patterns detected.