ClawHub

Video Workflow

Security checks across malware telemetry and agentic risk

Overview

This is a coherent cloud video-editing skill that discloses its remote processing, but users should only send media they are comfortable uploading to nemovideo.ai.

Install only if you trust nemovideo.ai with the clips, prompts, and generated media you provide. Use a dedicated token if possible, keep it private, and ask the agent to confirm before uploads, exports, or actions that may use credits when handling sensitive content.

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 (3)

Vague Triggers

Medium
Confidence
93% confidence
Finding
The getting-started prompt and example phrases are broad enough that ordinary conversation about editing or exporting video could invoke the skill without a clear, deliberate user opt-in. In this skill, unintended invocation is more concerning because it can initiate account/session setup and lead toward cloud upload and third-party processing of user media.

Vague Triggers

Medium
Confidence
96% confidence
Finding
The catch-all rule routing 'Everything else' to SSE gives the skill an ambiguous trigger scope, allowing many unrelated user prompts to be sent to the remote backend. Because SSE forwards free-form text to a cloud service and may alter session state, this broad routing increases the chance of unintended data disclosure and unauthorized remote actions.

Missing User Warnings

Medium
Confidence
95% confidence
Finding
The skill clearly instructs the agent to connect to a cloud backend, create tokens, maintain sessions, and upload/process user videos remotely, but it does not require a prominent user warning or explicit consent before sending media to a third-party service. This is especially sensitive because videos may contain personal, confidential, or regulated content, and the workflow also acquires anonymous tokens and persists session state on an external platform.

VirusTotal

62/62 vendors flagged this skill as clean.

View on VirusTotal