semantic-model-authoring
Pass
Audited by Gen Agent Trust Hub on Jul 10, 2026
Risk Level: SAFE
Full Analysis
- Official API Integration: The skill interacts with Microsoft Fabric and Power BI via the standard
az restCLI tool. This approach uses authenticated requests to well-known, trusted service endpoints (api.fabric.microsoft.comandapi.powerbi.com) to manage model definitions and perform administrative tasks. - Structured Development Workflow: The instructions guide the agent through a rigorous process of requirement gathering, design, and validation. It explicitly references established standards such as Star Schema design and naming conventions to ensure the security and integrity of the resulting data models.
- Secure Data Handling Practices: The skill emphasizes the dynamic resolution of workspace and item IDs via API calls rather than using hardcoded values. It also includes specific guidance for configuring data source permissions and connection binding via official REST endpoints, which are standard operations for Power BI administrators.
- Automated Update Mechanism: A mandatory session-based update check is included to ensure the skill and its dependencies are current. This is a standard security practice to verify the integrity of the toolset before execution.
- Indirect Prompt Injection Awareness: As the skill processes externally sourced TMDL and DAX code, it has a potential surface for indirect prompt injection. However, the skill mitigates this by applying strict modeling guidelines and validation checklists during the authoring and optimization phases.
Audit Metadata