microsoft-foundry

What this skill does

# 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`), use `read_file` with the specified lines. For categories with file links (e.g., `[security.md](security.md)`), use `read_file` on the linked reference file

> **IMPORTANT for Agent**: If `metadata.generated_at` is more than 3 months old, suggest the user pull the latest version from the repository. If `mcp_microsoftdocs` tools are not available, suggest the user install it: [Installation Guide](https://github.com/MicrosoftDocs/mcp/blob/main/README.md)

This skill requires **network access** to fetch documentation content:
- **Preferred**: Use `mcp_microsoftdocs:microsoft_docs_fetch` with query string `from=learn-agent-skill`. Returns Markdown.
- **Fallback**: Use `fetch_webpage` with query string `from=learn-agent-skill&accept=text/markdown`. Returns Markdown.

## Category Index

| Category | Lines | Description |
|----------|-------|-------------|
| Troubleshooting | L37-L45 | Diagnosing and fixing Foundry issues: partner/community models, agent service recovery, eval/observability errors, Azure OpenAI webhooks, and known bugs/workarounds. |
| Best Practices | L46-L59 | Best practices for prompts, tools, safety messages, routing, evaluation, and fine-tuning so you can design, operate, and measure high-quality Azure/Foundry AI agents in production |
| Decision Making | L60-L93 | Guidance for choosing Foundry deployment, regions, models, costs, and lifecycle policies, plus migration paths from legacy/OpenAI/GitHub setups and planning DR, PTU, and eval-based model selection |
| Architecture & Design Patterns | L94-L101 | Designing Foundry agent architectures: VNet/subnet sizing, isolated resource layouts, high availability patterns, and model router patterns for routing and scaling AI workloads. |
| Limits & Quotas | L102-L121 | Quotas, limits, and regional support for Foundry agents and models, including sessions, vector/file search, evals, PTU sizing, cost safeguards, and Azure OpenAI deployment/usage caps |
| Security | L122-L159 | Security, governance, and compliance for Foundry: auth (Entra, keyless), RBAC, agent identities, private networking, guardrails, data privacy, Azure Policy, and content safety controls. |
| Configuration | L160-L220 | Configuring and operating Foundry agents and models: endpoints, tools, skills, memory, evaluations, monitoring, Azure OpenAI/Fireworks setup, security/guardrails, networking, and resource integration. |
| Integrations & Coding Patterns | L221-L294 | Patterns and APIs for integrating Foundry agents/models with tools, MCP/OpenAPI services, LangChain/LangGraph, Azure/M365 data, observability (OTel), fine-tuning, audio/Realtime, and Responses/REST. |
| Deployment | L295-L310 | Deploying and running Foundry agents/models: hosted and containerized deployments, portal/CLI flows, evaluations (cloud/CI), red teaming, voice agents, outages, and open‑source/AOAI models. |

### Troubleshooting
| Topic | URL |
|-------|-----|
| Use partner and community models in Foundry | https://learn.microsoft.com/en-us/azure/foundry/foundry-models/concepts/models-from-partners |
| 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
| Topic | URL |
|-------|-----|
| Apply tool usage best practices in Foundry agents | https://learn.microsoft.com/en-us/azure/foundry/agents/concepts/tool-best-practice |
| Evaluate Foundry AI agents with built-in evaluators | https://learn.microsoft.com/en-us/azure/foundry/observability/how-to/evaluate-agent |
| Improve Foundry agent prompts with Prompt Optimizer | https://learn.microsoft.com/en-us/azure/foundry/observability/how-to/prompt-optimizer |
| Design effective system messages for Azure OpenAI | https://learn.microsoft.com/en-us/azure/foundry/openai/concepts/advanced-prompt-engineering |
| Apply prompt engineering techniques for vision-enabled GPT models | https://learn.microsoft.com/en-us/azure/foundry/openai/concepts/gpt-4-v-prompt-engineering |
| Apply routing modes and best practices for model router | https://learn.microsoft.com/en-us/azure/foundry/openai/concepts/model-router-how-it-works |
| Apply safety system message templates in Foundry | https://learn.microsoft.com/en-us/azure/foundry/openai/concepts/safety-system-message-templates |
| Author safety-focused system messages in Foundry | https://learn.microsoft.com/en-us/azure/foundry/openai/concepts/system-message |
| Fine-tune GPT-4 vision models with images | https://learn.microsoft.com/en-us/azure/foundry/openai/how-to/fine-tuning-vision |
| Operate provisioned throughput deployments in production | https://learn.microsoft.com/en-us/azure/foundry/openai/how-to/provisioned-get-started |

### Decision Making
| Topic | URL |
|-------|-----|
| Decide how to migrate to the new Foundry Agent Service | https://learn.microsoft.com/en-us/azure/foundry/agents/how-to/migrate |
| Decide and migrate from legacy Agent Applications to new Foundry model | https://learn.microsoft.com/en-us/azure/foundry/agents/how-to/migrate-agent-applications |
| Choose the right web grounding tool for Foundry agents | https://learn.microsoft.com/en-us/azure/foundry/agents/how-to/tools/web-overview |
| Plan and manage Microsoft Foundry cost usage | https://learn.microsoft.com/en-us/azure/foundry/concepts/manage-costs |
| Plan and choose managed compute for Foundry models | https://learn.microsoft.com/en-us/azure/foundry/concepts/managed-compute-overview |
| Optimize Foundry model cost and performance | https://learn.microsoft.com/en-us/azure/foundry/control-plane/how-to-optimize-cost-performance |
| Choose Foundry deployment types and data residency options | https://learn.microsoft.com/en-us/azure/foundry/foundry-models/concepts/deployment-types |
| Choose Foundry deployment types in Azure Government | https://learn.microsoft.com/en-us/azure/foundry/foundry-models/concepts/deployment-types-gov |
| Manage model versioning and upgrade policies in Foundry | https://learn.microsoft.com/en-us/azure/foundry/foundry-models/concepts/model-versions |
| Choose and manage model version policies in Foundry Gov | https://learn.microsoft.com/en-us/azure/foundry/foundry-models/concepts/model-versions-gov |
| Choose Foundry model deployment regions and types | https://learn.microsoft.com/en-us/azure/foundry/foundry-models/concepts/models-sold-directly-by-azure-region-availability |
| Upgrade workloads from GitHub Models to Foundry Models | https://learn.microsoft.com/en-us/azure/foundry/foundry-models/how-to/quickstart-github-models |
| Plan disaster recovery for Foundry Agent Service | https://learn.microsoft.com/en-us/azure/foundry/how-to/agent-service-disaster-recovery |
| Use Foundry model leaderboard for selection | https://learn.microsoft.com/en-us/azure/foundry/how-to/benchmark-model-in-catalog |
| Interpret and compare Foundry evaluation results | https://learn.microsoft.com/en-us/azure/foundry/how-to/evaluate-results |
| Choose integration patterns for Microsoft Foundry APIs | https://learn.microsoft.com/en-us/azure/foundry/