databricks-metric-views

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

• Third-party content exposure detected (high risk: 0.78). The required workflow for this skill (especially the Metric View Advisor) ingests outsider-authored free text at runtime from user-selected external Databricks assets—e.g., AI/BI dashboard serialized JSON (including widget labels/SQL text) and Genie space instructions/benchmark questions—then parses and uses that text to generate YAML/SQL that is fed back into the agent’s LLM context.