multi-database-literature-collector

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

• Third-party content exposure detected (high risk: 0.70). The skill's Execution steps — notably Step 2 (Database Selection) and Step 5 (Collect and Normalize Candidate Records) — explicitly require fetching and ingesting content from public third‑party sources (e.g., Google Scholar, bioRxiv/medRxiv and other preprint servers, Web of Science, ClinicalTrials.gov), and that untrusted/user‑generated content is read and used to assign priority tiers and guide next actions, so indirect prompt injection from those sources is possible.