ClawHub

VVVLINK Site Builder

Security checks across malware telemetry and agentic risk

Overview

This is an instruction-only website publishing skill whose network use, local credential storage, and public publishing behavior are disclosed and aligned with its purpose.

Install this only if you want an agent to create a VVVLink account credential, store it locally, send site-related content to VVVLink, and publish generated sites publicly. Review private, client, legal, or unreleased business content before using it, and be careful with delete or bulk cleanup requests because they can remove hosted site versions.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Tool MisuseTool Parameter Abuse, Chaining Abuse, Unsafe Defaults
  • 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
Findings (4)

Vague Triggers

Medium
Confidence
95% confidence
Finding
The trigger phrases are broad enough to activate on common conversations about websites or portfolios, which can unexpectedly route a user into a skill that creates accounts, stores credentials, writes files, and may publish content externally. In this skill, overbroad activation is more dangerous because the flow includes persistent side effects and external publication, increasing the chance of unintended execution.

Missing User Warnings

Medium
Confidence
98% confidence
Finding
The skill explicitly instructs the agent to publish immediately without confirmation, causing user-visible external changes on a third-party service. This is dangerous because a mistaken prompt interpretation or maliciously crafted content request could result in unintended public publication, account usage, and reputational harm.

Missing User Warnings

Medium
Confidence
94% confidence
Finding
The skill instructs the agent to send image search keywords derived from site content to an external service (`publish.vvvlink.com`) without requiring user consent or disclosure. If those keywords are based on private project details, client names, product plans, or other sensitive context, the agent may leak user data to a third party during normal operation.

Tool Parameter Abuse

High
Category
Tool Misuse
Content
---

## DELETE /sites/:siteId

Delete a site and all its versions.
Confidence
89% confidence
Finding
DELETE /sites/:siteId

VirusTotal

65/65 vendors flagged this skill as clean.

View on VirusTotal