ClawHub
Back to skill

Security audit

Livestream And Realtime

Security checks across malware telemetry and agentic risk

Overview

This skill is a disclosed livestream planning and promotion guide with clear limits around what the agent and WoopSocial can do.

Installers should treat this as an advisory social-content workflow. It may prepare and schedule public promotional or repurposed content through WoopSocial, so confirm platform, timing, copy, and clips before publishing; the live broadcast, moderation, shopping, and analytics remain human/native responsibilities.

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 (1)

Vague Triggers

Medium
Confidence
83% confidence
Finding
The eval includes very broad trigger phrases such as "Help me plan an Instagram Live," "Set up a livestream for me," and especially "Do my whole livestream end to end," which can cause the skill to activate for loosely related requests without strong gating. In an agentic system, over-broad invocation can lead to misrouting, unintended tool use, or the model asserting capabilities beyond the skill's intended advisory scope, particularly around live execution and scheduling boundaries.

VirusTotal

64/64 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.