adhoc-exploration-notebooks

Pass

Audited by Gen Agent Trust Hub on Mar 8, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill leverages official and well-known libraries including databricks-sdk, pyspark, pandas, and matplotlib, which are standard for data engineering.
  • [SAFE]: All external URLs point to official documentation sites for Databricks and Apache Spark.
  • [PROMPT_INJECTION]: The skill provides a surface for indirect prompt injection by design, as it allows for the exploration of external data tables. This is an inherent part of its functionality. * Ingestion points: assets/templates/exploration-notebook.py (via spark.table and spark.sql functions). * Boundary markers: None present in generated templates for table content. * Capability inventory: Access to Unity Catalog data and Databricks utilities (dbutils). * Sanitization: None performed on data retrieved from tables.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 8, 2026, 02:33 AM