Hosted Agents in Microsoft Foundry

Deploy and manage containerized AI agents on Microsoft Foundry with azd.

1|Updated Jan 14, 2026
One-click install
npx skills add https://github.com/EmeaAppGbb/spec2cloud-weather-agent --skill hosted-agents-in-microsoft-foundry
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: Hosted Agents in Microsoft Foundry
Source: https://github.com/EmeaAppGbb/spec2cloud-weather-agent/tree/main/.github/skills/hosted-agents
Command: npx skills add https://github.com/EmeaAppGbb/spec2cloud-weather-agent --skill hosted-agents-in-microsoft-foundry

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This Skill enables teams to deploy and manage containerized AI agents on Microsoft Foundry Agent Service, delivering enterprise-grade hosting, scaling, observability, and seamless Azure integration for production AI workflows.

Core Features & Use Cases

  • Hosted agent deployment: containerize AI agents and run them on Foundry with lifecycle management.
  • Observability and tracing: OpenTelemetry support, structured logging, and Azure Application Insights integration.
  • Lifecycle management and publishing: versioned deployments, scale adjustments, and publishing to Copilot/Teams with stable endpoints.
  • Tool integration: connect agents to MCP servers and Foundry tools for external data access.

Quick Start

Follow these steps to get started:

  1. Install and update Azure Developer CLI (azd) and log in to Azure.
  2. Define your hosted agent YAML and use azd ai agent to package, provision, and deploy to Foundry.
  3. Monitor deployment in Foundry portal and verify telemetry in Application Insights.

Frequently Asked Questions about Hosted Agents in Microsoft Foundry

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I deploy containerized AI agents to Microsoft Foundry?

You can deploy containerized AI agents to Microsoft Foundry by packaging a defined agent YAML file using the Azure Developer CLI (azd) to provision and manage the hosted runtime directly.

What is needed to host production AI workloads on Foundry?

Hosting production AI workloads on Foundry requires standard tooling such as azd and Docker, a defined agent YAML configuration, and an Azure subscription for scalable hosting and observability integration.

How does observability work for AI agents hosted on Azure Foundry?

Observability for AI agents hosted on Foundry works through native OpenTelemetry support, structured logging, and direct Azure Application Insights integration for tracing and monitoring production workflows.

Can I publish a deployed Foundry agent to Copilot or Teams?

Yes, you can publish deployed Foundry agents to Copilot and Teams using versioned lifecycle management, which provides stable endpoints to connect your hosted runtime to external platforms.

Does Foundry hosting support connecting agents to external data sources?

Yes, Foundry hosting supports connecting AI agents to external data sources by integrating with MCP servers and Foundry tools to access external data during production workflows.

What are the limitations of using azd for agent lifecycle management?

Using azd for agent lifecycle management requires a defined agent YAML and Docker containerization, meaning you must maintain standard tooling configurations to execute versioned deployments and scale adjustments properly.