How do I handle ACR permissions for hosted agents?

You must grant your Azure AI project's managed identity the 'AcrPull' role on the specific Azure Container Registry where your image is stored.

Which SDK version is required for this skill?

You must use azure-ai-projects version 2.0.0b3 or later to access the ImageBasedHostedAgentDefinition and hosted agent features.

What is an Azure AI hosted agent?

A hosted agent is a container-based AI assistant that runs within the Azure AI Foundry environment, allowing for custom environments and specialized resource allocations.

Can I use custom tools with these agents?

Yes, hosted agents support a variety of tools including the built-in Code Interpreter, File Search, and custom Model Context Protocol (MCP) servers.

Azure AI Hosted Agents (Python)

Name: Azure AI Hosted Agents (Python)
Author: sickn33

bysickn33

•

31,721

•

클라우드 인프라

Deploys and manages containerized AI agents on Azure AI Foundry using the Azure AI Projects Python SDK.

The Azure AI Hosted Agents skill enables developers to build and orchestrate enterprise-grade, container-based agents using the ImageBasedHostedAgentDefinition framework. This skill provides a standardized workflow for configuring resource allocations like CPU and memory, integrating Azure Container Registry (ACR), and implementing complex toolsets such as Code Interpreters and MCP servers. It is ideal for teams needing to move beyond basic LLM prompts into scalable, hosted agentic environments within the Azure AI Foundry ecosystem, ensuring secure authentication and robust version management.

주요 기능

0131,721 GitHub stars

02Granular resource allocation for CPU (0.5 to 4) and Memory (1Gi to 8Gi)

03Seamless integration with Azure Identity for secure authentication

04Comprehensive agent versioning and lifecycle management

05Support for multiple tool types including Code Interpreter and MCP

06Container-based agent deployment using Azure Container Registry (ACR)

사용 사례

01Deploying specialized data processing agents in isolated container environments

02Implementing custom agent protocols for enterprise-grade AI applications

03Scaling AI agent infrastructure with specific hardware requirements in Azure AI Foundry

주요 기능

0131,721 GitHub stars

02Granular resource allocation for CPU (0.5 to 4) and Memory (1Gi to 8Gi)

03Seamless integration with Azure Identity for secure authentication

04Comprehensive agent versioning and lifecycle management

05Support for multiple tool types including Code Interpreter and MCP

06Container-based agent deployment using Azure Container Registry (ACR)

사용 사례

01Deploying specialized data processing agents in isolated container environments

02Implementing custom agent protocols for enterprise-grade AI applications

03Scaling AI agent infrastructure with specific hardware requirements in Azure AI Foundry