supplementary-engagement

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

• Third-party content exposure detected (high risk: 0.85). The skill’s runtime workflow ingests the corpus’s text fields (including outsider-authored comment/body text) into the LLM context via the “Privacy Audit” and “Sentiment-Engagement Correlation” steps, where it scans and reports on that free-form content; this is outsider free text (e.g., comments/issues/threads authored by others) being read as prose for analysis/reporting.