querying-data-lake

Pass

Audited by Gen Agent Trust Hub on Sep 14, 2026

Risk Level: SAFE
Full Analysis
  • Destructive Statement Protection: The skill automatically classifies SQL statements and mandates user confirmation for INSERT, UPDATE, DELETE, and other destructive operations, preventing accidental data loss.
  • IAM-Aware Resource Selection: It implements sophisticated logic to select the appropriate Athena workgroup based on the caller's IAM ARN, ensuring queries run within the correct project context and reducing permission errors.
  • Cost and Performance Management: Instructions encourage the use of LIMIT clauses, partition filtering, and approximate functions for large datasets, helping to control scan costs and improve performance.
  • Tool Integrity: It enforces the use of verified AWS tools (MCP or official CLI) and validates caller identity before execution, ensuring the agent operates within the authorized AWS session.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 14, 2026, 11:26 PM
Security Audit — agent-trust-hub — querying-data-lake