patsnap-lifescience-target-intelligence

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

• Third-party content exposure detected (high risk: 1.00). The skill's mandatory workflow (see "Principle 5: Web Search Tool Usage" and PATH descriptions) instructs the agent to call web search and external public sources (e.g., ClinicalTrials.gov, PubMed/KEGG/uniprot and other web results) to fetch and ingest open third-party content which the agent must read and integrate into reports, creating a clear vector for indirect prompt injection.