ClawHub
Back to skill

Security audit

周易智慧顾问

Security checks for vulnerabilities and agentic risk

Overview

This skill is a local Zhouyi/I Ching report generator with some scoping and output-safety cautions, but no evidence of hidden data access, exfiltration, persistence, or destructive behavior.

Install only if you want an I Ching-style decision-reflection tool. Treat its reports as cultural or philosophical guidance, not medical, legal, financial, or life-determining advice. When generating reports, use a non-sensitive output filename and avoid opening reports made from untrusted question text.

SkillSpector

Vulnerability Patterns
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
  • 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
Findings (3)

Context-Inappropriate Capability

Medium
Confidence
96% confidence
Finding
The script accepts a user-controlled --output path, converts relative paths to the current working directory, and writes HTML directly to that location without any allowlist, sandbox, or path validation. In an agent/skill context this can overwrite arbitrary files writable by the process, which is unjustified for a divination/report-generation skill and becomes more dangerous because the generated HTML can also embed attacker-controlled question text.

Vague Triggers

Medium
Confidence
91% confidence
Finding
The trigger list includes broad phrases such as '帮我看看这件事', '给点建议', and '这个决定怎么样', which are common in ordinary conversation and can cause the skill to activate outside a clearly I Ching-related context. Overbroad activation can hijack unrelated user requests, causing unintended data collection, wrong-tool execution, or confusing behavior.

Vague Triggers

Medium
Confidence
89% confidence
Finding
The skill states it should activate on keywords or 'intent', but does not define clear boundaries for what constitutes that intent. Vague intent-based routing increases the chance of accidental invocation on loosely related discussions, especially because the skill is framed as giving decision support and generating reports.

Static analysis

No suspicious patterns detected.