creating-data-lake-table

Pass

Audited by Gen Agent Trust Hub on Jun 28, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill follows established practices for managing AWS resources using CLI tools. No evidence of obfuscation, persistence, or unauthorized privilege escalation was found.
  • [PROMPT_INJECTION]: The skill is designed to process external data, such as S3 file headers and user-provided descriptions, to infer database schemas (SKILL.md). This ingestion point is managed through a mandatory confirmation step where the agent must propose the schema to the user for approval before performing any write operations, which mitigates potential indirect prompt injection risks.
  • [COMMAND_EXECUTION]: The skill uses AWS CLI commands for infrastructure management, including aws s3tables and aws glue. These commands are consistent with the skill's primary function and include best practices like verifying credentials with aws sts get-caller-identity and checking for existing resources before creation.
  • [EXTERNAL_DOWNLOADS]: References to external repositories and documentation are limited to the official AWS GitHub organization and AWS documentation, which are trusted sources.
Audit Metadata
Risk Level
SAFE
Analyzed
Jun 28, 2026, 05:50 PM
Security Audit — agent-trust-hub — creating-data-lake-table