bigquery-migration-validator

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

• Third-party content exposure detected (low risk: 0.10). Customer-entered free text in the user's source database rows is read at runtime when the agent runs the row-level SELECTs on the mismatches_details and mismatches tables, but the agent must actively choose and run those queries, so outsider text is exposed only after an agent-initiated fetch rather than from a feed the workflow reads automatically.