data-engineering-storage-remote-access-integrations-iceberg

Pass

Audited by Gen Agent Trust Hub on Feb 16, 2026

Risk Level: LOW
Full Analysis
  • [Data Exposure] (SAFE): The skill uses appropriate placeholders for AWS credentials (AKIA...) and common default credentials for local development environments (minioadmin). It correctly recommends using environment variables or IAM roles for production security.
  • [External Dependencies] (SAFE): The skill depends on well-known and legitimate libraries: pyiceberg, pyarrow, and pandas. Installation commands use standard package managers.
  • [Indirect Prompt Injection] (LOW): The skill possesses a data ingestion surface but low risk due to structured data processing. 1. Ingestion points: table data read via catalog.load_table (SKILL.md). 2. Boundary markers: Absent. 3. Capability inventory: Storage modification via table.append and table.overwrite (SKILL.md). 4. Sanitization: Absent. Risk is low as data is treated as structured records rather than instructions.
Audit Metadata
Risk Level
LOW
Analyzed
Feb 16, 2026, 07:56 AM