impeccable

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

• Third-party content exposure detected (high risk: 0.80). The skill's critique and automated-detection workflow (see reference/critique.md and SKILL.md) explicitly requires opening and inspecting live pages and external URLs, injecting scripts into arbitrary web pages, and using those page contents/console outputs to drive the agent's analysis and recommended actions — meaning it fetches and interprets untrusted public web content as part of its runtime decisions.