full-page-screenshot

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

• Third-party content exposure detected (medium risk: 0.30). The skill’s --url and proxy fallback modes accept an outsider-provided --url and then use Chrome DevTools Protocol (Target.createTarget → waitForLoad → multiple Runtime.evaluate calls) to read/render the fetched page’s DOM content at runtime, which can include attacker-authored text.