Strategy Backtest

• Outcome: Pass
• Latest audit: May 29, 2026, 8:29 PM
• Version:
  1.0.0

Overview

This appears to be a legitimate local analysis helper, but it can install and run a local tool and save analysis history on the user's machine.

Install only if you are comfortable running a local Python CLI. Use read-only connector credentials, avoid storing sensitive experiment payloads unless needed, and periodically review or delete the local history database if it may contain confidential business data.

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 (2)

Lp3

Medium

• Category: MCP Least Privilege
• Confidence: 87% confidence
• Finding: The skill advertises no declared permissions, yet it appears capable of reading and writing local files. That mismatch undermines the permission model and can enable unexpected access to local data or persistence of artifacts without informed approval, especially in an agent environment where users rely on declared capabilities.

Tp4

High

• Category: MCP Tool Poisoning
• Confidence: 91% confidence
• Finding: The reported behavior goes beyond the stated purpose by persisting command history, migrating legacy local data, and returning static references unrelated to computation. Undisclosed persistence is risky because it can retain sensitive user inputs or analysis history, and behavior mismatch makes it harder for users and reviewers to assess what the skill actually does.

VirusTotal

65/65 vendors flagged this skill as clean.