ClawHub
Back to skill

Security audit

language-immersion-tv

Security checks across malware telemetry and agentic risk

Overview

This skill is a straightforward subtitle-analysis tool that reads user-chosen subtitle files and writes study-deck outputs without hidden network, credential, persistence, or destructive behavior.

Install only if you are comfortable letting the tool read subtitle files you point it at and write deck outputs to your chosen paths. Use authorized subtitle files and avoid running it over sensitive media libraries or private folders unless you intend those contents to be analyzed.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • MCP Least PrivilegeUnderdeclared Capability, Wildcard Permission, Missing Permission Declaration
  • 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)

Lp3

Medium
Category
MCP Least Privilege
Confidence
91% confidence
Finding
The skill clearly instructs users to read subtitle files and generate output files, which implies file_read and file_write capabilities, but it does not declare those permissions in metadata. Undeclared file access weakens transparency and permission scoping, making it easier for an agent runtime or reviewer to miss data-handling risks such as unintended access to local files or writing outputs into sensitive locations.

Missing User Warnings

Low
Confidence
84% confidence
Finding
The workflow encourages processing subtitle files from downloads and third-party media collections without any warning about privacy, licensing, or copyright sensitivity. While not a code-execution flaw, this omission can lead users to feed the skill personal or unlawfully obtained media-related files, creating legal and data-exposure risk in environments where file contents may be logged, shared, or retained.

VirusTotal

63/63 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.