azure-kusto-graph
Pass
Audited by Gen Agent Trust Hub on Aug 17, 2026
Risk Level: SAFECOMMAND_EXECUTION
Full Analysis
- Command Execution & File Writing: The skill includes procedures for the agent to write .kql files to the local workspace and launch the Kusto.Explorer.exe application. These actions are intended to streamline the developer workflow by moving queries from the chat interface to the native Windows client. The skill provides clear guidance on using ask_user for consent and warns against risky patterns when generating the associated PowerShell scripts.
- Indirect Prompt Injection Surface: As a tool that translates natural language and user-provided queries into complex KQL graph operations, the skill presents a surface for indirect prompt injection. However, because the resulting queries are displayed for user review and the execution (via kusto_query or local files) involves human-in-the-loop checkpoints, the risk is minimized.
Audit Metadata