Stock Monitor

• Outcome: Pass
• Latest audit: May 28, 2026 at 4:36 AM UTC
• Version:
  0.1.0

Overview

This skill is a disclosed stock-monitoring tool that polls financial data sources and can run as a user-started background monitor, with no evidence of credential theft, destructive actions, or hidden execution.

Install only if you are comfortable with a background process polling third-party financial sites for the stock codes and names in its watchlist. Treat its buy/sell/hold-style suggestions as informational alerts, not professional financial advice, and use the provided stop command when continuous monitoring is no longer needed.

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: 88% confidence
• Finding: The skill appears to require network access for market/news retrieval and continuous monitoring, but no permissions are declared in the manifest. Undeclared network capability weakens transparency and consent, making it harder for users or a hosting platform to understand what external access the skill will perform.

Tp4

High

• Category: MCP Tool Poisoning
• Confidence: 93% confidence
• Finding: The documented/observed behavior goes well beyond simple stock alerts into news scraping, sentiment analysis, capital-flow tracking, macro analysis, and investment advice. This scope expansion is dangerous because users may grant trust based on a narrow monitoring description while the skill performs broader data collection, external fetching, and advisory functions with different risk and compliance implications.

Description-Behavior Mismatch

Medium

• Confidence: 81% confidence
• Finding: The documentation markets the tool as an 'intelligent advisory system' with analysis and recommendations, while the manifest presents it as a monitoring/alert skill. This mismatch can mislead users about the degree of autonomy, external data use, and decision-support functionality, increasing the chance of unsafe reliance or unexpected behavior.

Description-Behavior Mismatch

Medium

• Confidence: 84% confidence
• Finding: The changelog advertises news and sentiment-analysis capabilities that are outside the stated seven-rule alert scope. Hidden or under-declared functionality is a security and trust issue because it suggests the effective behavior may exceed what users or reviewers expect from the manifest.

VirusTotal

64/64 vendors flagged this skill as clean.