deploy-model

To deploy an Azure OpenAI model using the CLI, provide a prompt like "deploy gpt-4o" to route execution to the appropriate deployment mode. The skill integrates with Azure CLI and REST APIs to verify authentication, select a project, and provision the deployment to Azure AI Foundry.

Intent-based routing for Azure OpenAI deployments analyzes user prompts to determine deployment intent and directs execution to Preset, Customize, or Capacity modes. This mechanism ensures accurate resource provisioning while aligning project scope and region capacity before deployment.

Yes, you need Azure CLI authentication to check Azure AI Foundry capacity. The skill validates Azure authentication and project scope via REST APIs before executing capacity discovery or deployment workflows to prevent misconfigurations and provisioning errors.

You can customize your Azure OpenAI deployment instead of using a preset by including specific intent keywords in your prompt. The skill routes to the Customize mode, allowing full configuration control while still enforcing guardrails and user confirmation before provisioning.

Prerequisites for deploying models to Azure AI Foundry include having Azure CLI, jq, python3, openssl, xxd, and base64 installed. These dependencies support authentication verification, project selection, and capacity checks before the deployment workflow executes.