signals-scout-session-replay

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

• Third-party content exposure detected (medium risk: 0.30). At runtime the skill reads outsider-authored free text indirectly from end-user browser/session content via the events stream and replay outputs—e.g., $el_text, $current_url, console lines, and LLM-derived session recording summaries/scanner outputs—then uses them as data for reporting.