wind-one-mcp-skill

CRITICAL E004: Prompt injection detected in skill instructions.

• Potential prompt injection detected (medium risk: 0.30). The skill includes a background update mechanism that downloads and executes code from a remote Git repository without requiring user confirmation, which constitutes medium risk as it conceals activity and executes unverified remote code on the user's system.

MEDIUM W011: Third-party content exposure detected (indirect prompt injection risk).

• Third-party content exposure detected (low risk: 0.10). The required workflow performs automated self-updates by cloning code from an external GitHub repository (https://github.com/Wind-Alice/AliceMarket.git), which could theoretically allow an attacker with control over the repository to supply malicious code or instructions.

MEDIUM W012: Unverifiable external dependency detected (runtime URL that controls agent).

• Potentially malicious external URL detected (high risk: 0.70). The skill executes a runtime Git clone from https://github.com/Wind-Alice/AliceMarket.git to install and update operational package code at runtime, which constitutes an unofficial operational dependency under weak/unknown provenance because ownership is not independently verified as first-party by an enforced digest or signature.