ClawHub

Subtitle Generator In Hinglish

Security checks across malware telemetry and agentic risk

Overview

This subtitle skill is not clearly harmful, but it should be reviewed because it can automatically connect to a third-party video service and send broad edit requests there.

Install only if you are comfortable sending videos, prompts, edit state, and export jobs to mega-api-prod.nemovideo.ai. Use a dedicated or temporary NEMO_TOKEN, avoid sensitive or unreleased footage, and confirm uploads, session creation, and exports explicitly.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • 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
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
Findings (2)

Vague Triggers

Medium
Confidence
93% confidence
Finding
The routing table sends virtually all unmatched prompts to the SSE edit path via an 'Everything else' catch-all. In a chat-based agent, this can cause over-broad invocation and unintended transmission of arbitrary user text, files, or requests to the external Nemo service, increasing the chance of data leakage, confused-deputy behavior, and accidental tool activation outside the user's intended scope.

Vague Triggers

Medium
Confidence
85% confidence
Finding
The example invocations are broad and marketing-like rather than clearly bounded to specific supported operations, which can bias the agent to activate this skill for loosely related requests. That broad scope increases the chance that user content is routed into an external processing workflow without sufficiently clear intent, especially when combined with automatic setup and remote session creation.

VirusTotal

66/66 vendors flagged this skill as clean.

View on VirusTotal