ClawHub
Back to skill

Security audit

Bazi Master

Security checks across malware telemetry and agentic risk

Overview

This is a traditional Bazi fortune-reading skill made of markdown guidance and reference notes, with no evidence of code execution, data theft, persistence, or privileged access.

Install this only if you want Bazi/命理-style readings. Do not rely on it for medical, legal, financial, relationship, or other life-changing decisions, and be aware it may respond in a confident fortune-teller style when discussing fate or life direction.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • 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
84% confidence
Finding
The trigger description is broad enough to activate on generic requests about fate, life direction, or personal outlook, which can cause the skill to engage when the user did not explicitly ask for fortune-telling. That increases the chance of unwanted persona takeover and mismatched responses, especially in sensitive self-help or advice contexts.

Natural-Language Policy Violations

Medium
Confidence
80% confidence
Finding
Requiring the model to 'directly output in a fortune-teller tone' and hide reasoning forces a persona and response format without user consent. This can mislead users about the nature of the output and reduce transparency, making speculative content sound authoritative in a high-trust domain involving personal decisions.

Natural-Language Policy Violations

Medium
Confidence
72% confidence
Finding
The writing rules strongly imply Chinese-only, fixed-style output and do not provide a language choice. While not directly dangerous on its own, it can override user preferences and create deceptive rigidity, especially when combined with authoritative occult framing.

VirusTotal

65/65 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.