ClawHub
Back to skill

Security audit

六维共振量化助手

Security checks for vulnerabilities and agentic risk

Overview

This is not malware, but it should be reviewed because it gives actionable stock-trading guidance while framing itself as only an analysis tool.

Install only if you deliberately want Chinese-language A-share stock analysis and understand it may produce trading-style recommendations. Treat any buy, sell, position-size, or stop-loss output as unverified decision support, not professional financial advice, and avoid automatic use on casual or ambiguous finance conversations.

SkillSpector

Vulnerability Patterns
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • MCP Tool PoisoningHidden Instructions, Unicode Deception, Parameter Description Injection
  • 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 (2)

Intent-Code Divergence

Medium
Confidence
95% confidence
Finding
The skill claims it is not a stock recommendation tool, but later explicitly instructs the assistant to provide buy/hold/sell advice, position sizing, and stop-loss guidance. This contradiction can mislead users and downstream systems about the skill's risk profile, causing the agent to deliver regulated or high-risk financial advice under a misleading disclaimer.

Vague Triggers

Medium
Confidence
89% confidence
Finding
The trigger description is broad enough to activate on many generic investing or learning queries, increasing the chance the skill is invoked outside its intended boundaries. In a financial context, over-triggering is risky because it can cause unsolicited or overly confident stock analysis to appear in situations where users did not explicitly request this specialized behavior.

Static analysis

No suspicious patterns detected.