ClawHub

work-critic

Security checks across malware telemetry and agentic risk

Overview

This is a text-only critique helper whose broad trigger phrases may activate too easily, but it does not show hidden access, persistence, credential use, or destructive behavior.

Install this if you want an agent to give direct, critical review feedback on submitted work. Consider invoking it explicitly or narrowing the triggers if you do not want ordinary words like review or critique to activate this mode.

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 (2)

Vague Triggers

Medium
Confidence
91% confidence
Finding
The trigger phrases include short, common expressions such as '请批评', '挑刺', 'review', and '弱点在哪里', which are broad enough to match ordinary conversation and unintentionally activate the skill. In an agent environment, this can cause prompt-routing mistakes, unexpected critique behavior, and accidental processing of unrelated or sensitive user content.

Vague Triggers

Medium
Confidence
95% confidence
Finding
The trigger list includes generic phrases such as “review”, “critique”, and short conversational prompts that can easily appear in ordinary user requests. This can cause unintended invocation of the skill, leading to prompt-routing errors, confusing behavior, or application of an unexpectedly harsh review persona in contexts where the user did not intend it.

VirusTotal

64/64 vendors flagged this skill as clean.

View on VirusTotal