asb-interview-learning

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

• Third-party content exposure detected (medium risk: 0.30). The skill’s runtime workflow ingests per-interview debrief files (including any pasted notes for interviews lacking a debrief file) and synthesizes over their “answers mapped to Q-numbers with addenda,” meaning outsider-authored free text in those debriefs is directly read and used for hypothesis updates.