ClawHub

life-restart-pro

Security checks across malware telemetry and agentic risk

Overview

This is a prompt-only entertainment skill for fictional life-simulation stories, with minor risk of broad trigger words causing accidental activation.

Install for casual fictional roleplay. Be aware that generic “restart” or “重开” language may invoke or rerun the simulator, and only approve file saving if you want a local archive of generated stories.

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
92% confidence
Finding
The activation phrases are very broad, including ordinary requests such as 'simulate my life' or 'what if I was born in XX,' which can cause the skill to trigger during unrelated conversations. Overbroad invocation increases the chance of unintended skill execution, response hijacking, and user confusion about which instructions are controlling the assistant.

Vague Triggers

Low
Confidence
88% confidence
Finding
The restart phrase includes generic terms like 'restart' and '重开', which are common in many contexts and may re-enter the skill unexpectedly. This can lead to accidental state resets or unwanted regeneration of content when a user is referring to something else.

Vague Triggers

Low
Confidence
90% confidence
Finding
The boundary handling allows generic reactivation phrases like '再来一次', '重开', and 'restart' without sufficient contextual constraints. In a multi-skill environment, these phrases can collide with normal user language or other skills, causing unintended reactivation and reducing isolation between behaviors.

VirusTotal

64/64 vendors flagged this skill as clean.

View on VirusTotal