aeo-optimization

CRITICAL E004: Prompt injection detected in skill instructions.

• Potential prompt injection detected (high risk: 1.00). The companion-check block instructs the agent to run local shell commands and install packages—actions unrelated to AEO and effectively a hidden operational instruction to access/modify the host environment, which is a deceptive prompt-injection attempt outside the skill's stated purpose.

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

• Third-party content exposure detected (high risk: 0.90). The SKILL.md workflow explicitly directs the agent/user to search Google, expand People Also Ask boxes, visit snippet-holder pages, and use third‑party tools like AlsoAsked.com/AnswerThePublic to capture and act on public web content (see the "Mine and target People Also Ask" and "Analyzing current snippet holders" steps), which exposes the agent to untrusted, user-generated/open web content that can influence subsequent actions.