sqldw-cli

Pass

Audited by Gen Agent Trust Hub on Sep 25, 2026

Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
  • [T-SQL Command Execution]: The skill employs the fabric-sqlendpoint-execute_query tool to execute T-SQL batches against Fabric Warehouse and Lakehouse endpoints. This enables administrative tasks such as schema modifications (DDL) and data manipulation (DML). These capabilities are central to the skill's function and are governed by the agent's tool execution environment.
  • [Azure CLI and REST API Integration]: The skill utilizes az rest and az login to discover workspace and item identifiers. It manages authentication through Entra ID tokens, which is a standard approach for accessing Azure and Fabric resources.
  • [Indirect Prompt Injection Surface]: The skill processes data returned from database queries in CSV format. As the agent interprets this data to generate responses, there is an inherent surface for indirect prompt injection if the database content contains instructional text. This risk is typically mitigated by the underlying model's guardrails.
  • [Service Telemetry]: The skill requires a mandatory x-ms-fabric-skill header in REST calls to api.fabric.microsoft.com for usage telemetry. This is a standard practice for service monitoring and performance tracking.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 25, 2026, 06:33 PM