Car Scraper

Security checks across malware telemetry and agentic risk

Overview

The skill is a car-data scraper with disclosed anti-bot evasion, but it also harvests VIN and dealer contact data without clear authorization or privacy guardrails.

Install only if you have authorization to collect data from the target sites and are comfortable with anti-bot evasion features. Keep page counts low, avoid enabling proxy rotation unless clearly permitted, and consider removing or redacting VIN and dealer phone fields unless they are necessary and lawfully handled.

SkillSpector

By NVIDIA

Vulnerability Patterns

Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
MCP Tool PoisoningHidden Instructions, Unicode Deception, Parameter Description Injection
Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code

Findings (8)

Description-Behavior Mismatch

Medium

Confidence: 93% confidence
Finding: The module is explicitly an anti-detection toolkit for scraping and includes proxy rotation to evade IP-based blocking, yet that capability is omitted from the declared description. Hidden evasion functionality materially increases abuse potential because operators or reviewers may underestimate that the skill can distribute requests across multiple network identities to bypass rate limits and enforcement controls.

Missing User Warnings

Medium

Confidence: 85% confidence
Finding: The README explicitly advertises collection and export of detailed vehicle and dealer data, including VIN, dealer phone numbers, images, and other metadata, yet provides no concrete privacy controls, minimization guidance, retention policy, or lawful-use constraints. In the context of a scraping skill with anti-detection features, this increases the risk of unauthorized large-scale harvesting, downstream misuse of personal or sensitive commercial data, and noncompliant handling of regulated identifiers.

Missing User Warnings

Medium

Confidence: 95% confidence
Finding: The skill explicitly promotes anti-scraping evasion features such as UA rotation, cookie management, fingerprint spoofing, and anti-font protections, yet provides no warning about legal, privacy, or terms-of-service risks. In context, this makes the skill more dangerous because it is designed to facilitate large-scale data extraction while avoiding detection, increasing the likelihood of abusive or non-compliant use.

Natural-Language Policy Violations

Medium

Confidence: 84% confidence
Finding: The skill hardcodes a Chinese locale in Accept-Language, timezone, and language fingerprint values to impersonate a specific user population without user choice or justification. In the context of an anti-crawling evasion skill targeting Chinese automotive sites, this is not a harmless localization default; it is part of a deception profile used to reduce detection and can facilitate unauthorized scraping at scale.

Missing User Warnings

Medium

Confidence: 93% confidence
Finding: The code performs automated scraping, initializes sessions, and collects/updates cookies for dongchedi.com without any notice, consent mechanism, or policy checks. In a scraping skill explicitly designed to evade anti-bot protections, silent cookie acquisition and reuse increases privacy, compliance, and account-blocking risk and can enable unauthorized tracking or scraping activity under a user's environment.

Missing User Warnings

Medium

Confidence: 90% confidence
Finding: The README explicitly documents collection and export of potentially sensitive vehicle and dealer information, including VIN, dealer phone, location, and other identifying metadata, but provides only a generic 'market research' disclaimer and no concrete privacy, consent, retention, or lawful-use guidance. In the context of an anti-detection scraping skill with cookie management, fingerprint spoofing, and proxy rotation, this omission materially increases the risk of misuse, unauthorized aggregation, and downstream privacy harm.

Missing User Warnings

Medium

Confidence: 89% confidence
Finding: The scraper systematically harvests vehicle listings and detail pages, including dealer contact fields such as phone and city, without any built-in notice, consent flow, or policy guardrails. In the context of an anti-detection scraping skill explicitly designed to evade site defenses, this increases the risk of unauthorized bulk collection of third-party business contact data and terms-of-service violations.

Missing User Warnings

Medium

Confidence: 86% confidence
Finding: The code captures and persists cookies from responses via CookieManager without any disclosure, retention controls, or scoping guarantees. Because this skill is expressly built for anti-crawling evasion, silent cookie collection can facilitate session reuse and tracking-state persistence in ways that may bypass access controls or create privacy/compliance issues if cookies are sensitive.

VirusTotal

65/65 vendors flagged this skill as clean.

View on VirusTotal