skills.rest
sohamda
sohamdaCOMMUNITY
Explore all skills by this author →
Website𝕏@iamsohamGitHub

microsoft-foundry

Deploy, evaluate, and manage Azure AI Foundry agents and model deployments end-to-end.

Updated May 12, 2026
Search Tags:#rbac#container registry#model deployment#agent deployment#quota management#azure cli#azure ai foundry
One-click install
npx skills add https://github.com/sohamda/apex-try-out-demo-repo --skill microsoft-foundry-sohamda
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: microsoft-foundry
Source: https://github.com/sohamda/apex-try-out-demo-repo/tree/main/.archive/_archived_skills/microsoft-foundry
Command: npx skills add https://github.com/sohamda/apex-try-out-demo-repo --skill microsoft-foundry-sohamda

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill requires azure-ai-projects, azure-identity, and includes references (resource) components.

What problem does it solve? Managing the full lifecycle of Azure AI Foundry agents requires coordinating many disconnected operations: provisioning projects and AI Services resources, deploying models with the right SKU and quota, containerizing and pushing agent code to ACR, invoking agents, running evaluations, and troubleshooting failures. This Skill orchestrates all of these workflows through structured sub-skills so nothing is missed. ## Core Features & Use Cases - Agent Lifecycle Management: Create hosted or prompt agents, containerize and push to ACR, deploy, invoke, and troubleshoot with validation after every step. - Model Deployment with Intelligent Routing: Routes requests to preset (quick deploy), customize (full control over version/SKU/capacity/RAI), or capacity discovery modes based on user intent. - Governance & Operations: Manage RBAC role assignments, quota and capacity planning, project connections, and evaluation datasets harvested from production traces. - Use Case: A developer asks to deploy a new LangGraph agent to Foundry. The Skill downloads the official sample, wraps it with the hosting adapter, builds a linux/amd64 Docker image, pushes to ACR, creates the hosted agent deployment, and validates it by sending a test message. ## Quick Start Ask the agent to deploy a new hosted agent to your Foundry project, and it will resolve your project context, provision what is missing, and validate each step before proceeding.

Frequently Asked Questions about microsoft-foundry

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

FAQPage Schema
How do I deploy a hosted agent to Azure AI Foundry?▼

Use the deploy sub-skill to containerize your agent code, build with --platform linux/amd64, push to Azure Container Registry, and create the hosted agent deployment. The workflow validates each step and routes failures to the troubleshoot sub-skill.

How do I deploy a GPT-4o model with a specific SKU and capacity?▼

The deploy-model sub-skill routes your request by intent: preset for quick defaults, customize for full control over version, SKU, capacity, and RAI policy, or capacity discovery to find regions with available quota before deploying.

What is the difference between prompt agents and hosted agents in Foundry?▼

Prompt agents (kind: prompt) are LLM-based agents backed by a model deployment and managed via MCP tools or the SDK. Hosted agents (kind: hosted) run your custom containerized code and require a hosting adapter exposing the OpenAI Responses API on port 8088.

Why does my Foundry model deployment fail with QuotaExceeded?▼

QuotaExceeded means your subscription's TPM or PTU limit in that region is fully consumed. Check regional usage with az rest against Microsoft.CognitiveServices usages, delete unused deployments to free quota immediately, or request an increase via the Azure Portal.

Can I convert an existing LangGraph or Agent Framework project to a Foundry hosted agent?▼

Yes. Add the appropriate hosting adapter package (azure-ai-agentserver-langgraph or azure-ai-agentserver-agentframework), wrap your agent as the default entrypoint, create an agent.yaml and Dockerfile exposing port 8088, then test locally before deploying.

When should I not use this Skill?▼

Do not use it for Azure Functions, App Service, or general Azure deployments, which belong to dedicated Azure deploy skills. It is scoped specifically to Microsoft Foundry resources, agents, models, quota, and RBAC.

Related Skills

Explore More Software Engineering Skills→
3horizons3horizons

ai-foundry-operations

Provision and manage Microsoft Foundry resources with Terraform H3 modules.

Official
Advanced
Supporter09Supporter09

langsmith-observability

Trace, evaluate, and monitor LLM application runs with LangSmith.

Community
Advanced
microsoftmicrosoft
1.4k

Azure App Onboard Scaffold

Generate Azure Bicep or Terraform templates from architecture plans.

Official
Advanced
Avihai-HAvihai-H

deploy-model

Deploy and configure Azure OpenAI models with Azure CLI and AI Foundry.

Community
Advanced