Expert guidance for Microsoft Foundry (Azure AI Foundry): building agents with Agent 365, MCP tools, Azure OpenAI, VNets, guardrails, and CI/CD. Requires the mcp_microsoftdocs MCP for live docs.
Install This Skill
npx skills add MicrosoftDocs/Agent-Skills# Download SKILL.md and place in your agent's skills folder
curl -o SKILL.md https://github.com/MicrosoftDocs/Agent-Skills/tree/main/skills/microsoft-foundry/raw/main/SKILL.md/skill add MicrosoftDocs/microsoft-foundrySKILL.md
Microsoft Foundry Skill
This skill provides expert guidance for Microsoft Foundry. Covers troubleshooting, best practices, decision making, architecture & design patterns, limits & quotas, security, configuration, integrations & coding patterns, and deployment. It combines local quick-reference content with remote documentation fetching capabilities.
How to Use This Skill
IMPORTANT for Agent: Use the Category Index below to locate relevant sections. For categories with line ranges (e.g.,
L35-L120), useread_filewith the specified lines. For categories with file links (e.g.,[security.md](security.md)), useread_fileon the linked reference file
IMPORTANT for Agent: If
metadata.generated_atis more than 3 months old, suggest the user pull the latest version from the repository. Ifmcp_microsoftdocstools are not available, suggest the user install it: Installation Guide
This skill requires network access to fetch documentation content:
- Preferred: Use
mcp_microsoftdocs:microsoft_docs_fetchwith query stringfrom=learn-agent-skill. Returns Markdown. - Fallback: Use
fetch_webpagewith query stringfrom=learn-agent-skill&accept=text/markdown. Returns Markdown.
Category Index
| Category | Lines | Description |
|---|---|---|
| Troubleshooting | L37-L44 | Diagnosing and fixing Foundry agent outages, data/resource loss, evaluation/observability issues, Azure OpenAI webhook failures, and known platform bugs with workarounds. |
| Best Practices | L45-L58 | Best practices for prompts, system messages, safety, tool use, routing, evaluation, and fine-tuning (incl. vision models) to improve Foundry/Azure AI agent quality. |
| Decision Making | L59-L97 | Guidance on choosing models, deployments, costs, and SDKs, plus migration, lifecycle, and disaster recovery decisions for Foundry, Gov, and related agent/model services. |
| Architecture & Design Patterns | L98-L105 | Designing Foundry agent architectures: VNet/subnet sizing, isolated deployments, high availability patterns, and spillover traffic strategies for provisioned workloads. |
| Limits & Quotas | L106-L126 | Limits, quotas, regions, and cost controls for Foundry agents and models, including sessions, vector/file search, function calls, evals, batch, caching, fine-tuning, and Azure OpenAI policies. |
| Security | L127-L161 | Security, privacy, and compliance for Foundry: auth/RBAC, agent identities, Agent 365 governance, network isolation, keys, policies/guardrails, content safety, and secure MCP/agent tooling. |
| Configuration | L162-L217 | Configuring and operating Foundry agents and models: endpoints, tools, memory, workflows, security/guardrails, networking, monitoring, Azure OpenAI, and deployment/quotas. |
| Integrations & Coding Patterns | L218-L271 | Integrating Foundry agents and models with tools, APIs, and external systems (MCP, LangChain, LangGraph, Azure services, OpenAI features), plus coding patterns for routing, tracing, and fine-tuning. |
| Deployment | L272-L284 | Deploying and running models and agents in Foundry: portal/CLI/Bicep deployment, hosted agent migration and recovery, CI/CD evaluations, red teaming scans, and custom/fine-tuned model hosting. |
Troubleshooting
| Topic | URL |
|---|---|
| Recover Foundry Agent Service from resource and data loss | https://learn.microsoft.com/en-us/azure/foundry/how-to/agent-service-operator-disaster-recovery |
| Troubleshoot Foundry evaluation and observability issues | https://learn.microsoft.com/en-us/azure/foundry/observability/how-to/troubleshooting |
| Set up and troubleshoot Azure OpenAI webhooks | https://learn.microsoft.com/en-us/azure/foundry/openai/how-to/webhooks |
| Known issues and workarounds for Microsoft Foundry services | https://learn.microsoft.com/en-us/azure/foundry/reference/foundry-known-issues |
Best Practices
Decision Making
Architecture & Design Patterns
| Topic | URL |
|---|---|
| Design networking and subnet sizing for Foundry agents | https://learn.microsoft.com/en-us/azure/foundry/agents/concepts/agents-networking-deep-dive |
| Plan standard agent setup with isolated resources | https://learn.microsoft.com/en-us/azure/foundry/agents/concepts/standard-agent-setup |
| Design high availability for Microsoft Foundry agents | https://learn.microsoft.com/en-us/azure/foundry/how-to/high-availability-resiliency |
| Design spillover traffic management for provisioned deployments | https://learn.microsoft.com/en-us/azure/foundry/openai/how-to/spillover-traffic-management |
Limits & Quotas
Security
Configuration
Integrations & Coding Patterns
Deployment
Related Skills
Quick Stats
Source repo · this pathAuthor
Microsoft
@MicrosoftDocs