fabric-lakehouse

Pass

Audited by Gen Agent Trust Hub on Sep 15, 2026

Risk Level: SAFE
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill provides templates for reading external data files (CSV, JSON, Parquet) and Delta tables from Lakehouse storage into Spark DataFrames. While this creates a surface for processing untrusted data, the templates follow standard data engineering practices.
  • [DYNAMIC_EXECUTION]: Includes various PySpark and Spark SQL code snippets for data transformation, merging, and maintenance. These are intended for use within a secure Spark environment and do not involve unsafe execution of untrusted code.
  • [DATA_EXFILTRATION]: Mentions integration with external storage providers like Amazon S3 and Google Cloud Storage through Fabric Shortcuts. These are documented as standard platform features for cross-cloud data access and do not contain exfiltration commands.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 15, 2026, 06:51 AM
Security Audit — agent-trust-hub — fabric-lakehouse