deploy-model
Pass
Audited by Gen Agent Trust Hub on Mar 23, 2026
Risk Level: SAFE
Full Analysis
- Project Selection and Confirmation: The skill implements mandatory confirmation steps before any deployment occurs. This ensures that the agent verifies the target project and region with the user, preventing accidental resource creation in incorrect environments.
- Dynamic Resource Discovery: The skill avoids hardcoded lists of models or SKUs. Instead, it dynamically queries official Azure APIs (via
az cognitiveservices model list) to determine available versions and capabilities in real-time, ensuring accuracy and reliability. - Capacity and Quota Validation: The capacity discovery sub-skill and pre-deployment phases perform checks against both platform-wide capacity and the user's specific subscription quota limits. This provides the user with clear information on where deployments are possible before execution.
- Secure Command Execution: Interactions with Azure resources are conducted through the Azure CLI (
az) and Resource Manager REST APIs (az rest). These operations target well-known, trusted endpoints (e.g.,management.azure.com) and rely on the user's existing authenticated session. - Data Processing via Local Scripts: The skill utilizes local shell and PowerShell scripts for complex data transformation and ranking. These scripts process JSON data returned from official Azure tools using standard utilities like
jqand inline Python, which is a standard pattern for managing Azure infrastructure.
Audit Metadata