Stock Monitor

• Outcome: Review
• Latest audit: May 28, 2026, 11:43 AM
• Version:
  1.3.0

Overview

This appears to be a real stock alert skill, but it includes an undocumented Futu/OpenD monitor with password-based access that goes beyond the advertised Yahoo Finance checks.

Review before installing, especially if you expected only Yahoo Finance price alerts. Install only if you are comfortable with scheduled background checks, local portfolio-related state files, Yahoo Finance requests, and the included but under-documented Futu/OpenD script. Do not place brokerage or Futu unlock passwords in the config unless the publisher clearly documents why they are needed, how they are protected, and how to disable or remove that mode.

SkillSpector

By NVIDIA

Vulnerability Patterns

• MCP Least Privilege: Underdeclared Capability, Wildcard Permission, Missing Permission Declaration
• MCP Tool Poisoning: Hidden Instructions, Unicode Deception, Parameter Description Injection
• Prompt Injection: Instruction Override, Hidden Instructions, Exfiltration Commands
• Data Exfiltration: External Transmission, Env Variable Harvesting, File System Enumeration
• Privilege Escalation: Excessive Permissions, Sudo/Root Execution, Credential Access

Findings (4)

Lp3

Medium

• Category: MCP Least Privilege
• Confidence: 89% confidence
• Finding: The skill documentation instructs the agent to read and write files under the user's home directory and to fetch stock data over the network, but the skill declares no permissions. This creates a transparency and policy-enforcement gap: users and the platform cannot accurately assess or constrain what the skill will access, which is especially relevant because it stores persistent state and reaches external data sources.

Tp4

High

• Category: MCP Tool Poisoning
• Confidence: 95% confidence
• Finding: The documented behavior says the skill uses Yahoo Finance for simple threshold-based monitoring, but the finding indicates the implementation also uses Futu OpenD/futu APIs, can unlock a market connection with an unlock_password, and supports broader monitoring logic than disclosed. This mismatch is dangerous because it can cause users to provide or expose brokerage-related credentials and grant trust under false assumptions about external connections, data handling, and operational scope.

Description-Behavior Mismatch

Medium

• Confidence: 95% confidence
• Finding: The implementation uses the Futu OpenD API while the skill metadata claims Yahoo Finance, creating a trust and transparency mismatch. This is dangerous because users may expose local trading infrastructure, credentials, or network access they would not expect from the advertised functionality, especially since Futu OpenD is a locally connected brokerage-related service rather than a public market-data API.

Intent-Code Divergence

Medium

• Confidence: 93% confidence
• Finding: The module docstring explicitly contradicts the manifest by stating that it uses Futu Open API. Such inconsistencies are a supply-chain and user-consent risk because reviewers and users may approve the skill based on one trust model while the code operates under another, potentially granting broader local-system or financial-service access than expected.

VirusTotal

66/66 vendors flagged this skill as clean.