Agent-Skills microsoft-foundry
Expert knowledge for Microsoft Foundry (aka Azure AI Foundry) development including troubleshooting, best practices, decision making, architecture & design patterns, limits & quotas, security, configuration, integrations & coding patterns, and deployment. Use when building Foundry agents with Azure OpenAI, tools, vector stores, RAG, CI/CD deployments, or M365/Teams apps, and other Microsoft Foundry related development tasks. Not for Microsoft Foundry Classic (use microsoft-foundry-classic), Microsoft Foundry Local (use microsoft-foundry-local), Microsoft Foundry Tools (use microsoft-foundry-tools).
git clone https://github.com/MicrosoftDocs/Agent-Skills
T=$(mktemp -d) && git clone --depth=1 https://github.com/MicrosoftDocs/Agent-Skills "$T" && mkdir -p ~/.claude/skills && cp -r "$T/skills/microsoft-foundry" ~/.claude/skills/microsoftdocs-agent-skills-microsoft-foundry && rm -rf "$T"
skills/microsoft-foundry/SKILL.mdMicrosoft 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.,
), useL35-L120with the specified lines. For categories with file links (e.g.,read_file), use[security.md](security.md)on the linked reference fileread_file
IMPORTANT for Agent: If
is more than 3 months old, suggest the user pull the latest version from the repository. Ifmetadata.generated_attools are not available, suggest the user install it: Installation Guidemcp_microsoftdocs
This skill requires network access to fetch documentation content:
- Preferred: Use
with query stringmcp_microsoftdocs:microsoft_docs_fetch
. Returns Markdown.from=learn-agent-skill - Fallback: Use
with query stringfetch_webpage
. Returns Markdown.from=learn-agent-skill&accept=text/markdown
Category Index
| Category | Lines | Description |
|---|---|---|
| Troubleshooting | L37-L41 | Known Microsoft Foundry bugs, error codes, limitations, and their workarounds, plus steps to diagnose and resolve common platform and deployment issues. |
| Best Practices | L42-L53 | Best practices for configuring tools, prompts, system messages, evaluation, fine-tuning (incl. vision), and optimizing performance/latency for Azure OpenAI/Foundry agents |
| Decision Making | L54-L86 | Guidance on choosing and upgrading Foundry models, tools, and deployments, planning migrations (Azure OpenAI, GitHub Models, Assistants), and optimizing cost, performance, and versioning strategies. |
| Architecture & Design Patterns | L87-L98 | Architectural patterns for Foundry agents: standard setup, RAG/indexing, HA/DR, regional recovery, provisioned throughput, spillover traffic, and LLM routing optimization. |
| Limits & Quotas | L99-L112 | Quotas, rate limits, regions, and cost controls for Foundry agents, models, vector stores, evals, batch jobs, prompt caching, Azure OpenAI usage, and limited-access policies. |
| Security | L113-L145 | Security, privacy, and governance for Foundry: auth/RBAC, keys and encryption, network isolation, guardrails/safety, compliance, and secure use of tools, agents, and model connections |
| Configuration | L146-L193 | Configuring Foundry agents, models, tools, storage, monitoring, security/guardrails, and Azure OpenAI integrations, including workflows, tracing, evaluations, and external resource setup. |
| Integrations & Coding Patterns | L194-L261 | Patterns and APIs for integrating Foundry agents and models with tools, gateways, LangChain/LangGraph, Azure OpenAI, search, speech, image, tracing, security, and external services. |
| Deployment | L262-L278 | Deploying agents and models: infra setup, Bicep/CLI/portal deployment, publishing to M365/Teams/Azure, lifecycle management, CI/CD evaluations, regions, and custom/fine-tuned models. |
Troubleshooting
| Topic | URL |
|---|---|
| Troubleshoot Microsoft Foundry known issues and workarounds | https://learn.microsoft.com/en-us/azure/foundry/reference/foundry-known-issues |