lamindb

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

• Third-party content exposure detected (high risk: 0.70). The skill explicitly instructs importing and using public third‑party data (e.g., bt.CellType.import_source() in Use Case 1) and lists HTTP/HTTPS endpoints and HuggingFace datasets in Integrations, so the agent will fetch and interpret public web/third‑party content that can affect annotation and downstream actions.