cb-analytics-schema

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

• Third-party content exposure detected (high risk: 0.70). The skill’s infer_schema(dataset, sample_size, cluster) runtime path reads up to sample_size documents from the specified dataset and summarizes their top-level fields, so the LLM context can include outsider-authored free text contained in those documents (e.g., data ingested by other users/tenants).