ClawHub
Back to skill

Security audit

taste.skill

Security checks across malware telemetry and agentic risk

Overview

This is a non-executable paper-review guidance skill, with the main caveat that it can activate automatically for paper-triage tasks.

Install this if you want a strict paper-triage assistant. Be aware it may activate automatically in paper-review contexts, may search for related work when tools are available, and may produce a severe Chinese-language review style unless the user gives clearer preferences.

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
88% confidence
Finding
The skill explicitly enables implicit invocation, which can cause the agent to activate this skill without a clear, user-directed opt-in. That creates a real security and reliability risk because strict triage behavior could be triggered in broader conversations unexpectedly, potentially influencing outputs or exposing user content to a skill pathway the user did not intend to use.

VirusTotal

62/62 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.