novelty-gap-check

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

• Third-party content exposure detected (high risk: 0.75). The skill’s runtime LLM context is populated by the user’s idea description (operating-user authored) and by retrieval results from superpower-writing:research-lookup / superpower-writing:literature-review (which fetch and ingest outsider-authored web/arXiv/abstract text), so it can include outsider free text via the prior-work search pipeline.