MrBeast.skill - 内容创造操作系统

Security checks across malware telemetry and agentic risk

Overview

This appears to be a prompt-only YouTube strategy/persona skill with no evidence of code execution, credential access, data exfiltration, or persistence.

Install if you want YouTube strategy advice in a strong MrBeast-inspired voice. Be aware that the persona may dominate tone, and consider invoking it explicitly only when you want that style; do not treat the persona as real attribution or expert authority.

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 (2)

Natural-Language Policy Violations

Low
Confidence
92% confidence
Finding
The skill explicitly forces the assistant into a first-person MrBeast persona and suppresses normal framing after the first turn. This can mislead users about source/authority, reduce transparency, and make it harder for the system to preserve user-preferred tone or safe defaults, though it does not directly enable code execution or data exfiltration.

Vague Triggers

Medium
Confidence
89% confidence
Finding
The activation rules are broad enough to trigger on ordinary YouTube optimization questions like CTR, thumbnails, hooks, and retention, even when the user did not request this specific skill. That can cause unintended routing into a strong persona-driven mode, increasing the chance of confused-deputy behavior, hidden prompt takeover of general queries, and responses that do not match user intent.

VirusTotal

65/65 vendors flagged this skill as clean.

View on VirusTotal