cross-domain-analogy-finder

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 includes “Search the abstraction…”, “Provenance… Ground prior-art through literature-triangulation”, and “Every DEFENSE claim needs a real, checkable source”, which implies fetching/reading outsider-authored web or API-delivered text (e.g., prior-art pages/papers/issues) and ingesting it into the LLM context for classification and provenance.