ClawHub
Back to skill

Security audit

master-huineng

Security checks across malware telemetry and agentic risk

Overview

This is a topic-focused Huineng/Chan teaching skill with disclosed citation and style rules, limited optional web lookup, and no persistence or privileged system access.

Install this if you want a Chan/Zen and Huineng-focused teaching voice with strict source citation. Be aware it may activate broadly on Zen-related terms and may use FoJin web lookup for source verification when local excerpts are insufficient.

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
90% confidence
Finding
The trigger criteria are intentionally expansive and include invoking the skill whenever a question 'touches' Chan/Zen doctrine, even without an explicit request. This can cause unintended routing, override user expectations, and increase the chance that unrelated conversations are steered into a constrained religious/persona response path.

Natural-Language Policy Violations

Medium
Confidence
88% confidence
Finding
The skill mandates a specific persona and speaking style ('direct,' Huineng-like voice, prescribed openings) rather than making it optional based on user consent. That can misrepresent the assistant's role, create unwanted religious or authoritative framing, and reduce user agency by forcing roleplay or doctrinal tone onto ordinary informational queries.

VirusTotal

63/63 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.