skills.rest
mordechai30
mordechai30COMMUNITY
Explore all skills by this author →
GitHub

microsoft-foundry

Orchestrate Azure AI Foundry agent lifecycle with Azure Developer CLI.

Updated Aug 27, 2026
Search Tags:#deployment#llm#evaluation#fine-tuning#agents#foundry#azd
One-click install
npx skills add https://github.com/mordechai30/.agents --skill microsoft-foundry-mordechai30
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: microsoft-foundry
Source: https://github.com/mordechai30/.agents/tree/main/skills/microsoft-foundry
Command: npx skills add https://github.com/mordechai30/.agents --skill microsoft-foundry-mordechai30

SYSTEM DOCUMENTATION & REQUIREMENTS

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

What problem does it solve?

This skill addresses the complexity of managing the entire AI agent lifecycle, from initial infrastructure provisioning and model deployment to continuous evaluation, fine-tuning, and production troubleshooting.

Core Features & Use Cases

  • Full Lifecycle Orchestration: Manage agent scaffolding, deployment, and CI/CD pipelines using Azure Developer CLI (azd) and Foundry tools.
  • Evaluation & Optimization: Run batch evaluations, harvest production traces for datasets, and optimize prompts or agent instructions to improve performance.
  • Model Customization: Perform SFT, DPO, and RFT fine-tuning with integrated grader calibration and training curve analysis.
  • Use Case: A developer can use this skill to provision a secure Foundry project, deploy a hosted agent, monitor its performance via App Insights, and fine-tune the underlying model based on production trace data.

Quick Start

Use the microsoft-foundry skill to scaffold and deploy a new hosted agent end-to-end by following the quick-start-hosted workflow.

Frequently Asked Questions about microsoft-foundry

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

FAQPage Schema
How do I manage the end-to-end lifecycle for Azure AI Foundry agents?▼

You can manage the Azure AI Foundry agent lifecycle by orchestrating infrastructure provisioning, model deployment, evaluation, and fine-tuning. This skill uses Azure Developer CLI and Foundry SDKs to handle agent scaffolding, CI/CD pipelines, and production telemetry analysis.

What is the best way to fine-tune an LLM using production trace data?▼

The best way to fine-tune an LLM with production trace data is to harvest telemetry traces to build datasets, then apply SFT, DPO, or RFT fine-tuning. This process includes integrated grader calibration and training curve analysis to optimize model performance.

Do I need the Azure Developer CLI to deploy and evaluate AI agents?▼

Yes, you need the Azure Developer CLI (azd) along with the azure-ai-projects and azure-identity Python packages. These dependencies are required to provision secure Foundry projects, manage agentic resources, and execute evaluation suites.

Can I run batch evaluations and optimize prompts for hosted AI agents?▼

Yes, you can run batch evaluations and optimize prompts for hosted AI agents. The skill allows you to harvest production traces for datasets and refine agent instructions to systematically improve evaluation scores and overall agent performance.

How does infrastructure provisioning work for Azure AI Foundry projects?▼

Infrastructure provisioning for Foundry projects works by using the Azure Developer CLI to scaffold and secure resources. This establishes the necessary environment to deploy hosted agents and monitor their performance via Application Insights telemetry.

Related Skills

Explore More Software Engineering Skills→
microsoftmicrosoft
242

microsoft-foundry

Discover, deploy, and manage Microsoft Foundry AI resources with Python SDK and MCP tools.

Official
Advanced
forgefabrikforgefabrik

finetuning

Fine-tune Azure AI Foundry models with SFT, DPO, or RFT jobs.

Community
Advanced
qprjack86qprjack86

microsoft-foundry

Manage Azure AI Foundry agent and model workflows from deployment through evaluation.

Community
Advanced
anishi1222anishi1222

microsoft-foundry

Manage Foundry agent lifecycle including deployment, evaluation, fine-tuning, and monitoring.

Community
Advanced