ClawHub

Web Navigator

Security checks across malware telemetry and agentic risk

Overview

This appears to be an unfinished template rather than a working web-navigation skill, with no malicious code or privileged behavior shown.

Before installing, understand that this likely will not provide useful web-navigation behavior yet. Prefer a completed version that explains when it activates, how it handles untrusted web content, and what actions it may take.

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 (2)

Vague Triggers

Medium
Confidence
96% confidence
Finding
The manifest description is still a TODO placeholder and does not define when this skill should be invoked or what boundaries it operates within. In an agent system, vague or missing trigger conditions can cause mis-selection of the skill, unintended use in inappropriate contexts, and unsafe downstream behavior because neither users nor orchestration logic have clear constraints.

Vague Triggers

Medium
Confidence
98% confidence
Finding
The body of the skill is only template scaffolding and contains no actual operational guidance, constraints, or safe usage instructions. For a web-navigation skill, this is risky because browsing-related capabilities can reach untrusted content; without concrete procedures and guardrails, an agent may improvise behavior, mishandle external input, or operate beyond intended scope.

VirusTotal

66/66 vendors flagged this skill as clean.

View on VirusTotal