ClawHub
Back to skill

Security audit

Zhugeliang Think Partner

Security checks across malware telemetry and agentic risk

Overview

This is a text-only strategy-advice skill with no code execution, data access, persistence, or hidden install behavior.

Install only if you want a Chinese-first, Zhuge Liang-themed strategy and planning lens. It appears safe from a security perspective, but its broad triggers may shape generic planning conversations more than expected.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
Findings (3)

Vague Triggers

Medium
Confidence
93% confidence
Finding
The activation section lists many common, high-level prompts such as planning, prioritization, and resource constraints without defining exclusion conditions or clear trigger boundaries. In an agent environment, this can cause the skill to activate for ordinary conversations too broadly, leading to prompt hijacking of user intent, unwanted persona injection, or interference with more appropriate domain-specific skills.

Vague Triggers

Medium
Confidence
91% confidence
Finding
The description uses broad need-based language like 'when you need strategic planning' and 'complex situation analysis,' which overlaps with a large fraction of normal assistant interactions. This increases the chance of over-activation, causing the skill to inappropriately shape responses, crowd out user preference, or misroute tasks that should remain generic or be handled by specialized skills.

Natural-Language Policy Violations

Medium
Confidence
84% confidence
Finding
Declaring Chinese as the primary language without stating that it should follow the user's language preference can override expected interaction settings and reduce usability or transparency. While not directly enabling code execution or data exfiltration, it can still create consent and control issues by forcing outputs into a language the user did not request.

VirusTotal

64/64 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.