verify-citation

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

• Third-party content exposure detected (high risk: 0.85). The skill fetches outsider-authored free text from public web APIs at runtime—specifically CrossRef JSON metadata and OpenAlex abstract text via urllib.request.urlopen() in crossref_by_doi() / crossref_search() / openalex_abstract(), which are then inserted into the LLM-visible report output (including the abstract).