figure-pipeline

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

• Third-party content exposure detected (high risk: 0.90). The skill's Step 1 (RETRIEVE) explicitly fetches reference figures and captions from open public sources (e.g., PubMed/eutils, bioRxiv API, Europe PMC) and then instructs the agent to extract and use those references to plan, style, and generate figures, so untrusted third-party content could materially influence actions.