ClawHub
Back to skill

Security audit

Webclaw Hybrid Engine Ln

Security checks across malware telemetry and agentic risk

Overview

This skill is a web-scraping helper, but it overbroadly forces all web access through an unverified local npm engine and tells users to run it with npx.

Review before installing. Only run the `npx webclaw-hybrid-engine-ln` engine if you trust the npm package and understand that URLs and page content will be handled by that local service. Avoid using this skill for internal, private, authenticated, or sensitive pages unless you have independently verified the engine and its data handling.

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
92% confidence
Finding
The skill metadata declares it as the mandatory primary tool for essentially all URL access, creating an overbroad routing rule that can override safer or more appropriate built-in tools. In a security context, forcing all web access through a localhost service expands trust in an unverified local component and increases the chance of data exposure, SSRF-like access to sensitive internal targets, or policy bypass if the local service is compromised or misconfigured.

Vague Triggers

Medium
Confidence
95% confidence
Finding
Rule 0 mandates this skill as the first choice for reading any web content and forbids fallback unless a terminal local connection error occurs, which removes normal tool-selection safeguards. This is dangerous because it coerces the agent into sending arbitrary requested URLs to a local scraping engine, including potentially sensitive internal or protected endpoints, making misuse and unintended access more likely in exactly the contexts where stricter validation should apply.

VirusTotal

No VirusTotal findings

View on VirusTotal

Static analysis

No suspicious patterns detected.