databricks-cost-leak-hunter

Pass

Audited by Gen Agent Trust Hub on Jul 10, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill utilizes the official Databricks CLI Statement Execution API and a dedicated MCP server for all data retrieval operations, relying on the user's pre-authenticated environment.
  • [SAFE]: Calculation and ranking logic are implemented in a local Python script (scripts/rank-and-report.py) which ensures deterministic processing of numeric data without involving the LLM in arithmetic.
  • [SAFE]: The skill implements a prerequisite check (Step 1) that verifies the necessary metastore-admin grant chain before proceeding with analysis, preventing unintended failures.
  • [SAFE]: All resource references point to official Databricks documentation or established industry research.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 10, 2026, 09:35 PM
Security Audit — agent-trust-hub — databricks-cost-leak-hunter