design-inspector

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

• Third-party content exposure detected (high risk: 1.00). The skill autonomously navigates to and evaluates any user-provided public URL (see "browser_navigate → $ARGUMENTS" in Step 1 and the multiple browser_evaluate snippets in Steps 2 and 5 of SKILL.md), ingesting and interpreting live third‑party webpage content which can materially influence subsequent actions like clicks/hovers and report generation.