azure-kusto-graph

Pass

Audited by Gen Agent Trust Hub on Aug 10, 2026

Risk Level: SAFE
Full Analysis
  • [Secure Command Handling]: The skill provides clear instructions for handling database management commands that modify or delete metadata (e.g., .drop, .create-or-alter). It mandates that the agent displays the full command and obtains explicit user confirmation via ask_user before execution.
  • [Safe File Generation]: For the optional Kusto Explorer integration, the skill outlines a secure procedure for creating .kql files on the local system. It specifically recommends avoiding unsafe coding patterns like PowerShell here-strings to prevent potential command injection and requires user consent for file creation.
  • [Local Application Integration]: The skill includes a mechanism for opening local vendor applications on Windows. This feature is gated by an explicit user approval process and targets specific, known executables within the user's environment.
  • [Structured Query Generation]: By translating natural language into standard KQL graph patterns, the skill promotes consistent and predictable query generation, reducing the risk of unintended database interactions.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 10, 2026, 06:30 PM
Security Audit — agent-trust-hub — azure-kusto-graph