ClawHub

Zyte API

Security checks across malware telemetry and agentic risk

Overview

This is a coherent Zyte web-scraping helper that uses ClawLink for credentials and tool access, with some normal privacy and account-scope considerations.

Install only if you are comfortable enabling the ClawLink plugin and connecting a Zyte API account. Treat any URL, request parameters, and extracted page content as data that may be processed by ClawLink and Zyte; avoid secrets-bearing links, internal systems, private pages, regulated data, or high-volume scraping unless you have authorization and understand the cost and terms-of-service impact.

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
78% confidence
Finding
The activation text is broad enough to match generic scraping, extraction, or monitoring requests, which can cause the skill to be invoked in situations where users did not clearly consent to sending target URLs and retrieved content to an external third-party service. In an agent ecosystem, over-broad routing increases the risk of unintended data disclosure and unnecessary use of powerful external collection capabilities.

Missing User Warnings

Medium
Confidence
91% confidence
Finding
The skill instructs users to extract and scrape websites through ClawLink and Zyte without prominently warning that requested URLs and returned page content are transmitted to external services. This creates a meaningful risk of unintended disclosure when users provide private URLs, sensitive query parameters, internal endpoints, or confidential page contents under the assumption that processing is local.

VirusTotal

57/57 vendors flagged this skill as clean.

View on VirusTotal