negative-news-triage

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

• Third-party content exposure detected (high risk: 0.75). The skill’s runtime workflow ingests “hit set” content that is derived from public web/open-source pulls and vendor adverse-media exports (outsider-authored text), which the LLM then summarizes into the per-hit assessment and memo fields (e.g., headline summaries, source trace, typology/materiality), creating indirect prompt-injection exposure via that outsider text.