ClawHub

Shows

Security checks across malware telemetry and agentic risk

Overview

This is a coherent show-tracking skill that stores local entertainment notes, with some broad trigger wording users should understand before enabling.

Install if you are comfortable with the agent keeping entertainment and family viewing notes in ~/shows/. Use explicit commands for saving progress or recommendations, and avoid recording sensitive family details unless you want them stored locally.

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 (3)

Vague Triggers

Medium
Confidence
85% confidence
Finding
The trigger logic is broad enough that ordinary conversation about watching media could activate the skill and cause unintended writes or status changes in `~/shows/`. Because this skill stores personal viewing history and preferences, accidental activation can corrupt records or capture data the user did not intend to log.

Vague Triggers

Medium
Confidence
89% confidence
Finding
The trigger phrase "recommend something" is broad enough to match ordinary conversation, which can cause the skill to activate outside the user's clear intent to use this specific movie/TV tracking function. In an agent setting, overly broad activation can lead to unintended access to personal viewing history, watchlists, or recommendation workflows when the user was speaking generically.

Vague Triggers

Medium
Confidence
84% confidence
Finding
The condition "When someone recommends something" is ambiguous about speaker identity and context, so the skill may capture and store recommendations from incidental conversation, quoted text, or third-party content without clear user consent. That creates privacy and integrity risks because the agent could incorrectly memorialize social interactions or external content as user-tracked data.

VirusTotal

64/64 vendors flagged this skill as clean.

View on VirusTotal