ClawHub
Back to skill

Security audit

yanmu-hot-stocks

Security checks across malware telemetry and agentic risk

Overview

This skill is not malicious, but it needs review because it presents stock picks as dynamic market data while the script only returns fixed sample lists.

Treat this as a teaching/demo skill, not a source of current market recommendations. Before installing, expect fixed sample stock lists unless the publisher implements real data fetching and clearly labels source and freshness in the output.

SkillSpector

By NVIDIA
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 (3)

Description-Behavior Mismatch

Medium
Confidence
97% confidence
Finding
The skill advertises dynamic hot-stock retrieval, but the implementation only returns a hardcoded fallback list. In a stock-recommendation context, this can mislead users into believing they are receiving current market-driven information when they are actually seeing static, potentially stale recommendations, which can influence financial decisions.

Intent-Code Divergence

Medium
Confidence
95% confidence
Finding
The helper functions are documented as fetching live data from Sina/Yahoo Finance but always return None, creating a deceptive implementation gap between documented behavior and actual behavior. In a financial-analysis skill, this increases the risk that downstream users or components trust nonexistent live-data collection and make decisions based on static fallback content.

Vague Triggers

Medium
Confidence
94% confidence
Finding
The skill states it is 'automatically invoked' without defining clear user intent boundaries, trigger conditions, or approval requirements. In an agent environment, broad auto-invocation can cause the model to run external data-fetching logic unnecessarily, increasing the chance of unintended tool execution, network access, or user confusion about when the skill is acting on their behalf.

VirusTotal

64/64 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.