databricks-spark-structured-streaming

Pass

Audited by Gen Agent Trust Hub on Jul 31, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill provides comprehensive instructional content and code examples for Spark Structured Streaming. Analysis across all 10 threat categories confirms that the code follows established security best practices for the Databricks environment.
  • [SAFE]: External dependencies are restricted to well-known, industry-standard libraries such as psycopg (for PostgreSQL connectivity), pyspark, and delta-spark.
  • [SAFE]: Secret management is correctly handled using dbutils.secrets.get(), ensuring that no credentials (passwords, API keys, or SASL JAAS configurations) are hardcoded in the templates.
  • [SAFE]: Data persistence and checkpointing are configured using Unity Catalog Volumes and persistent cloud storage (S3/ADLS), avoiding insecure or ephemeral storage locations.
  • [SAFE]: Custom database sinks (e.g., in references/lakebase-sink-python.md) implement parameterized queries and standard database drivers, mitigating common command injection risks in the context of streaming ETL tasks.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 31, 2026, 08:24 AM
Security Audit — agent-trust-hub — databricks-spark-structured-streaming