ClawHub
Back to skill

Security audit

三层筛网量化选股v1.0

Security checks across malware telemetry and agentic risk

Overview

The skill does the advertised stock screening, but it embeds and automatically uses a third-party data-service login.

Review before installing. The main issue is not the stock-screening code itself, but the exposed JQData credential: it should be removed, rotated, and replaced with user-provided credentials through environment variables or secure configuration. Treat outputs as informational screening results only, and only enable recurring runs if you intentionally want scheduled finance scans.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • 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
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
Findings (4)

Context-Inappropriate Capability

High
Confidence
99% confidence
Finding
The file hardcodes live JQData credentials and immediately uses them for authentication at import time. This exposes a third-party account to anyone who can read, copy, or run the skill, enabling unauthorized use, quota exhaustion, account abuse, and possible access to data or billing tied to that account.

Vague Triggers

Medium
Confidence
84% confidence
Finding
The trigger language is broad enough to match generic stock-related requests, which can cause this finance skill to activate outside its intended quantitative-screening context. In a financial domain, over-broad activation is risky because it may produce stock recommendations or screening outputs when the user wanted general education, market commentary, or a narrower task, increasing the chance of inappropriate financial guidance.

Vague Triggers

Medium
Confidence
86% confidence
Finding
The usage examples include unconstrained phrases like generating stock recommendation lists, which may encourage invocation for broad investment-advice scenarios rather than a bounded screening workflow. Because the skill operates in a sensitive finance context, ambiguous examples increase the chance that an agent uses it as a recommendation engine without proper guardrails, disclosures, or user confirmation.

Missing User Warnings

High
Confidence
99% confidence
Finding
Hard-coded credentials for an external data provider are a real secret-management vulnerability. In this skill context, the danger is heightened because the code is meant to be distributed and reused, so the embedded account can be harvested from the file or triggered automatically by importing the module, causing unauthorized service access and operational abuse.

VirusTotal

60/60 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.