nimble-databricks-data-products

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

• Third-party content exposure detected (high risk: 0.85). The skill ingests outsider-authored free text from public web sources at runtime via nimble_integration.tools.nimble_agent_run(...) (called in Phase 3’s correlated LATERAL join), and then stores/exposes it as readable raw VARIANT in the unified results table, which can later be used in dashboard/app queries.