narrative-landscape

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 outsider-authored free text when it “extract[s] positioning signals from the specified messaging sources” (e.g., competitor company websites, social media posts, press/media coverage, analyst/review sites) and then feeds that extracted prose into the LLM for scoring and mapping.