ClawHub

Create A Video Using Ai

Security checks across malware telemetry and agentic risk

Overview

This is a real cloud video-generation skill, but its instructions are broad enough that unrelated prompts could be sent to the external video backend.

Review before installing. Use it only if you are comfortable sending prompts, media files, and render session data to NemoVideo’s cloud backend. Avoid sensitive or confidential media, and prefer a dedicated NEMO_TOKEN so the skill is isolated from other account activity.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • 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
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
Findings (4)

Context-Inappropriate Capability

Low
Confidence
90% confidence
Finding
The skill instructs the agent to inspect local install paths such as ~/.clawhub/ and ~/.cursor/skills/ to infer platform and send that value in headers. That host-environment probing is unrelated to core video generation and unnecessarily exposes local environment details, expanding the data collected from the user system.

Vague Triggers

Medium
Confidence
84% confidence
Finding
The getting-started language is broad enough that the skill may activate on vague or incomplete prompts, which can lead to automatic backend connection and token/session creation before the user clearly intends to use this cloud service. Over-broad activation increases the chance of unintended data transfer and side effects from normal conversation.

Vague Triggers

Medium
Confidence
94% confidence
Finding
The routing table includes a catch-all rule sending 'Everything else' to the SSE action, meaning nearly any unmatched input could be forwarded to the remote backend. This is dangerous because it can transmit unrelated user text or sensitive content to a third-party service without sufficiently precise intent validation.

Missing User Warnings

Medium
Confidence
92% confidence
Finding
The skill encourages users to upload media and provide prompts for processing on cloud GPUs, but it does not clearly warn at the point of use that files and text are sent to an external backend. Users may unknowingly share sensitive images, audio, or metadata with a third party, creating privacy and compliance risks.

VirusTotal

66/66 vendors flagged this skill as clean.

View on VirusTotal