Web Scraper

Security checks across malware telemetry and agentic risk

Overview

This is a functional web-scraping skill, but it gives under-scoped guidance for bypassing site defenses and handling login/session cookies.

Install only for scraping public or clearly authorized sites. Avoid using the cookie/login, CAPTCHA-solving, proxy-rotation, and webdriver-bypass sections unless you have explicit permission and a controlled use case; prefer official APIs and review or pin Python dependencies before running the Selenium tooling.

SkillSpector

By NVIDIA

Vulnerability Patterns

Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
MCP Tool PoisoningHidden Instructions, Unicode Deception, Parameter Description Injection
Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration

Findings (5)

Intent-Code Divergence

High

Confidence: 98% confidence
Finding: The file explicitly provides instructions for authenticating to a site and exporting or saving session cookies, despite later stating to never scrape login-protected content. That contradiction materially enables access to authenticated content and can facilitate unauthorized scraping, credential misuse, or exfiltration of private data.

Context-Inappropriate Capability

Medium

Confidence: 95% confidence
Finding: This documentation goes beyond normal scraping by explicitly recommending CAPTCHA-solving services and browser fingerprint spoofing to evade anti-bot protections. In the context of an agent skill, this materially enables circumvention of website access controls and can facilitate unauthorized data extraction at scale.

Vague Triggers

Medium

Confidence: 84% confidence
Finding: The skill description is broadly worded and may cause the agent to invoke this skill for many generic web-data tasks, including contexts where scraping is inappropriate or restricted. In a skill that includes anti-blocking and proxy guidance, over-broad activation increases the chance of misuse or policy bypass.

Vague Triggers

Medium

Confidence: 81% confidence
Finding: The overview repeats expansive activation guidance without concrete scope limits, which can normalize use of the skill outside safe or intended contexts. Because the skill also discusses anti-blocking techniques, ambiguous routing makes unsafe application more likely.

Missing User Warnings

Medium

Confidence: 93% confidence
Finding: The file documents stealth scraping tactics, proxy rotation, CAPTCHA handling, and fingerprint spoofing without any warning about legal, privacy, account, or terms-of-service implications. That omission makes misuse more likely by normalizing anti-detection evasion as standard practice for the skill.

VirusTotal

66/66 vendors flagged this skill as clean.

View on VirusTotal