microsoft-foundry:microsoft-foundry

Azure AI Foundry agent deployment and troubleshooting are managed through a comprehensive suite of tools that handle the complete agent lifecycle, including creation, deployment, invocation, and issue resolution for both prompt-based and hosted agents.

Managing Azure AI Foundry model capacity and quotas involves discovering available capacity, optimizing quota usage, and handling deployment limits to ensure efficient resource allocation for your hosted AI models.

Yes, you can deploy custom AI models in Azure AI Foundry with full control over versions, SKUs, capacity discovery, and Responsible AI policies using deterministic operations orchestrated via Azure CLI and MCP tools.

Yes, Azure CLI is required to configure RBAC for AI Foundry resources, as the Skill orchestrates complex workflows for capacity management, model deployment, and access control using Azure CLI and MCP tools.

Azure AI Foundry hosted agent invocation failures can be diagnosed using the Skill's built-in troubleshooting and observability features, which help identify issues across the agent lifecycle, environment variables, and deployment configurations.

Yes, this approach works for configuring environment variables and monitoring production AI agents by providing observability features that allow you to test agent responses and monitor their performance in a production environment.