ClawHub
Back to skill

Security audit

Web Scraping

Security checks across malware telemetry and agentic risk

Overview

This appears to be a web-scraping skill, but it can work on logged-in pages and save collected data without enough user-control guidance.

Install only if you intentionally want an agent to scrape sites, including potentially logged-in pages. Before use, confirm the target site, fields to collect, whether account/private data may be accessed, and where results will be stored; avoid using it for sensitive accounts unless you explicitly need that data and can manage retention and site-policy risk.

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 skill description is broad enough to match many ordinary user requests involving websites, extraction, monitoring, or summaries, which can cause over-invocation of this skill outside narrowly intended scraping scenarios. In practice, that increases the chance the agent will use browser automation or persistent workspace storage on requests where the user did not clearly intend scraping behavior, creating unnecessary privacy, compliance, and data-handling risk.

Missing User Warnings

Medium
Confidence
91% confidence
Finding
The skill explicitly supports login-gated scraping and saving results to the workspace, but it does not require warning the user about credentialed access, terms-of-service restrictions, sensitive data capture, or local retention of scraped content. That omission is dangerous because the agent may collect account-scoped or personal data and persist it without explicit consent or minimization, increasing privacy, legal, and confidentiality exposure.

VirusTotal

64/64 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.