ClawHub
Back to skill

Security audit

web-snippet-extractor

Security checks across malware telemetry and agentic risk

Overview

This skill coherently extracts code snippets from user-chosen public web pages and does not show hidden or destructive behavior.

Before installing, be aware that it fetches external web pages and may save extracted snippets if you ask it to. Review snippets from untrusted pages before reusing them, especially shell commands or configuration involving secrets.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
Findings (2)

Vague Triggers

Medium
Confidence
86% confidence
Finding
The activation criteria are broad enough to trigger on common web-browsing and documentation tasks, which can cause the skill to run in many contexts beyond the user's explicit intent. Over-broad auto-invocation increases the chance of fetching untrusted pages and surfacing adversarial snippets or instructions, expanding the attack surface for prompt injection and unsafe downstream use.

Missing User Warnings

Low
Confidence
93% confidence
Finding
The skill instructs writing extracted content to workspace files without requiring explicit user confirmation or warning that local data will be modified. Even though the destination path is constrained, this can still cause unexpected persistence of untrusted web content, clutter local state, or facilitate indirect prompt-injection persistence in later workflows.

VirusTotal

64/64 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.