ClawHub
Back to skill

Security audit

浏览器CLI工具-专业版

Security checks across malware telemetry and agentic risk

Overview

This browser automation skill is broadly coherent, but it asks for account automation, session-state reuse, persistent scheduling, screenshots, and webhook notifications without enough scoping or user-control detail.

Review before installing. Use this only for explicit browser automation tasks, verify the npm package source, avoid shared login-state stores unless your team has access controls, and require confirmation before account actions, screenshots, form submissions, scheduled jobs, or webhook notifications.

SkillSpector

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
92% confidence
Finding
The skill claims command execution should be restricted and user input should not be interpolated into command lines, yet elsewhere documents callback URLs, webhooks, external API integration, and command execution features without enforceable constraints. In an exec-capable skill focused on browser automation, this mismatch can enable data exfiltration to attacker-controlled endpoints or unsafe command/network behavior if downstream implementations trust user-supplied destinations.

Vague Triggers

Medium
Confidence
84% confidence
Finding
The trigger condition is so broad that it can cause the skill to activate for many generic productivity or automation requests, increasing the chance it is used in inappropriate or higher-risk contexts without informed consent. Because the skill includes browser automation, batch actions, screenshots, persistent state, and exec, over-triggering materially raises the risk of unintended account actions or data handling.

Missing User Warnings

Medium
Confidence
95% confidence
Finding
The skill documents batch sign-in, automated form submission, screenshot capture, shared state storage, and webhook notifications, but does not provide clear user-facing warnings about account impact, personal data processing, credential handling, or transmission to external services. In this context, the absence of explicit safeguards makes misuse and accidental privacy/security harm significantly more likely.

VirusTotal

64/64 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.