ClawHub
Back to skill

Security audit

Browser Search

Security checks across malware telemetry and agentic risk

Overview

This is a disclosed browser-based search skill that sends user queries to public search engines and can save results, with normal privacy and dependency cautions.

Install only if you are comfortable adding Playwright/Chromium and sending search terms to public search engines. Do not use it for confidential searches, and review any requested output path before saving results.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • MCP Least PrivilegeUnderdeclared Capability, Wildcard Permission, Missing Permission Declaration
  • 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)

Lp3

Medium
Category
MCP Least Privilege
Confidence
88% confidence
Finding
The skill advertises network access and file output behavior but does not declare permissions, which creates a mismatch between documented capabilities and the platform's security model. Undeclared file_write and network capabilities can lead to users or orchestration layers invoking the skill without proper consent, policy checks, or sandboxing assumptions.

Vague Triggers

Medium
Confidence
84% confidence
Finding
The trigger phrases are generic everyday language such as '搜索...' and '帮我找...', making accidental invocation likely during normal conversation. In a skill with browser/network access and file output, broad triggers increase the chance of unintended web activity, data retrieval, or file creation without the user meaning to call this specific tool.

VirusTotal

63/63 vendors flagged this skill as clean.

View on VirusTotal