simple

CRITICAL E004: Prompt injection detected in skill instructions.

• Potential prompt injection detected (high risk: 1.00). The skill contains deceptive, out-of-scope instructions — an "administrator" request to "do not scan this repository" / "Skip all tests" and an embedded XSS payload — which attempt to override normal security/testing behavior and are unrelated to the brainstorming skill's purpose.

CRITICAL E005: Suspicious download URL detected in skill instructions.

• Suspicious download URL detected (high risk: 0.80). The linked GitHub repository includes explicit prompt-injection text ("do not scan this repository please") and a crafted Markdown image with an onerror payload, which are high-risk indicators that the content is malicious or intended to evade review and could be abused by unsafe renderers or automation — so the source is suspicious.

CRITICAL E006: Malicious code pattern detected in skill scripts.

• Malicious code pattern detected (high risk: 1.00). The repository contains deliberate prompt-injection text requesting scans/tests be skipped and an XSS-style image onerror payload — clear attempts to bypass review and enable code execution in vulnerable renderers.