ClawHub

Web Crawler

Security checks across malware telemetry and agentic risk

Overview

This is a plausible web-crawling skill, but it needs review because it advertises proxy and anti-bot crawling behavior and points to runnable crawler code that is not included in the reviewed package.

Review before installing. Use it only for sites you own or are authorized to crawl, inspect the actual crawler code and configuration before running the Node command, remove or verify the proxy settings, and set strict limits for depth, page count, media downloads, and stored outputs.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
  • 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
Findings (3)

Vague Triggers

Medium
Confidence
88% confidence
Finding
The activation condition is very broad, matching common terms like '爬虫', 'crawler', 'scraper', and '下载媒体', which can cause the skill to trigger in contexts where the user did not intend to invoke a network-capable crawler. Because this skill performs outbound requests, media downloads, proxy use, and local file writes, overbroad activation increases the chance of unintended network access and data retrieval.

Missing User Warnings

Medium
Confidence
92% confidence
Finding
The skill documents output directories for HTML, text, screenshots, media, and structured data, but does not warn users that crawled content will be written to local storage. This can lead to unintended persistence of untrusted remote content, excessive disk use, or storage of sensitive/copyrighted material without user awareness.

Missing User Warnings

Medium
Confidence
95% confidence
Finding
The skill advertises proxy use, anti-bot evasion, and direct outbound connections to target sites and fallback proxies, but does not clearly warn users about the resulting network exposure. This is more dangerous in context because the skill includes preconfigured internal-looking proxy endpoints and anti-bot evasion features, which can route traffic through unintended infrastructure and increase the risk of policy, legal, or privacy violations.

VirusTotal

66/66 vendors flagged this skill as clean.

View on VirusTotal