Skip to main content
skill CX Codex

microsoft-extensions-ai — Claude Code skill: what it does

Skill for building provider-agnostic .NET AI integrations with Microsoft.Extensions.AI: chat clients, embeddings, tools, structured output, and evaluation.

MIT ★ 521 updated 2026-10-29

Install into Claude Code · Skills CLI (Vercel) All methods and platforms ↓
npx skills add managedcode/dotnet-skills --skill microsoft-extensions-ai -a claude-code
Open on GitHub

What microsoft-extensions-ai does

This skill guides the agent through building and reviewing code that uses Microsoft.Extensions.AI, the provider-agnostic abstraction layer for .NET AI applications. It covers IChatClient and IEmbeddingGenerator composition in DI, middleware for caching, logging, and telemetry, tool calling via AIFunction, vector search composition, and evaluation. The agent reads repository context, edits files, and runs build and test commands.

The skill distinguishes between plain model access, tool calling, embeddings, image generation, local models via Ollama, and full agent orchestration. For the latter, it escalates to microsoft-agent-framework. It treats chat state explicitly: for stateless providers, it resends history; for stateful ones, it propagates ConversationId. When working with vectors, the skill treats the embedding model, dimensions, and collection schema as a single contract: changing any of them requires reindexing.

Tool descriptions consume request tokens, so the skill registers only the tools relevant to the current conversation. For RAG, it wires Microsoft.Extensions.VectorData and Microsoft.Extensions.DataIngestion instead of hand-rolled store wrappers.

What it helps with

  • write a chat client without locking into one AI provider
  • add tool calling and structured output to a .NET AI application
  • set up middleware for caching, logging, and telemetry around chat
  • build a RAG pipeline with vector search and embeddings
  • test AI integration with a local Ollama model

Example prompts for the agent

  • Add IChatClient to DI with caching and logging, use OpenAI provider
  • Register a .NET method as AIFunction for model invocation
  • Build a RAG pipeline: read documents, chunk them, generate embeddings, write to VectorStore

How to install microsoft-extensions-ai

1Skills CLI (Vercel)

npx skills add managedcode/dotnet-skills --skill microsoft-extensions-ai -a claude-code

2Manually: put the files in place

~/.claude/skills/microsoft-extensions-ai/SKILL.md

Copy the whole skill folder; for one project use .claude/skills/microsoft-extensions-ai/

Step by step, with commands for each platform: How to install skills in Claude Code: five ways

Questions about microsoft-extensions-ai

How to install microsoft-extensions-ai in Claude Code?

Run npx skills add managedcode/dotnet-skills --skill microsoft-extensions-ai -a claude-code. The skill will appear in ~/.claude/skills/microsoft-extensions-ai/SKILL.md.

What does the microsoft-extensions-ai skill do?

It helps build and review code that uses Microsoft.Extensions.AI: chat clients, embeddings, tool calling, structured output, middleware, vector search, and evaluation. The agent edits files, runs builds, and runs tests.

When should I use microsoft-agent-framework instead of this skill?

When you need multi-agent orchestration, agent threads, durable execution, or remote agent hosting. For provider-agnostic chat, embeddings, and tools, Microsoft.Extensions.AI is enough.

Source text

SKILL.md · MIT
---
name: microsoft-extensions-ai
description: "Build provider-agnostic .NET AI integrations with `Microsoft.Extensions.AI`, `IChatClient`, embeddings, middleware, structured output, vector search, and evaluation. USE FOR: building or reviewing .NET code that uses Microsoft.Extensions.AI, Microsoft.Extensions.AI.Abstractions, IChatClient, IEmbeddingGenerator, ChatOptions, or AIFunction;. DO NOT USE FOR: unrelated stacks; generic tasks that do not need this specific guidance. INVOKES: inspect the repository context, edit targeted files, and run relevant build, test, lint, or validation commands when changes are made."
compatibility: "Requires `Microsoft.Extensions.AI` or a .NET AI application that needs model, embedding, tool-calling, or evaluation composition without full agent orchestration."
---

# Microsoft.Extensions.AI

## Trigger On

- building or reviewing `.NET` code that uses `Microsoft.Extensions.AI`, `Microsoft.Extensions.AI.Abstractions`, `IChatClient`, `IEmbeddingGenerator`, `ChatOptions`, or `AIFunction`
- adding `IImageGenerator`, local-model chat via Ollama, AI app templates, or the `.NET AI` quickstarts for assistants and MCP
- choosing between low-level AI abstractions, provider SDKs, vector-search composition, evaluation libraries, and a fuller agent framework
- adding streaming chat, structured output, embeddings, tool calling, telemetry, caching, or DI-based AI middleware
- wiring `Microsoft.Extensions.VectorData`, `Microsoft.Extensions.DataIngestion`, MCP tooling, or evaluation packages around a provider-agnostic AI app

## Workflow

1. Classify the request first: plain model access, tool calling, embeddings/vector search, evaluation, image generation, local-model prototyping, MCP bootstrap, or true agent orchestration.
2. Default to `Microsoft.Extensions.AI` for application and service code that needs provider-agnostic chat, embeddings, middleware, structured output, and testability.
3. Reference `Microsoft.Extensions.AI.Abstractions` directly only when authoring provider libraries or lower-level reusable integration packages.
4. Model `IChatClient` and `IEmbeddingGenerator` composition explicitly in DI. Keep options, caching, telemetry, logging, and tool invocation inspectable in the pipeline.
5. Treat chat state deliberately. For stateless providers, resend history. For stateful providers, propagate `ConversationId` rather than assuming all providers behave the same way.
6. Use `Microsoft.Extensions.VectorData` and `Microsoft.Extensions.DataIngestion` as adjacent building blocks for RAG instead of hand-rolling store abstractions prematurely. Treat the embedding model, vector dimensions, and collection schema as one owned contract: changing any of them means reindexing rather than reusing old vector data. Keep vector API source-breaking notes version-aware; in the 10.5+ line, named-argument usage of `VectorStoreVectorAttribute` uses `dimensions:`.
7. Treat the `.NET AI` quickstarts as bootstrap paths, not finished architecture. They now cover minimal assistants, MCP client/server flows, local models, app templates, and image generation. Start there for a vertical slice, then harden the DI, telemetry, and evaluation story here.
8. Escalate to `microsoft-agent-framework` when the requirement becomes agent threads, multi-agent orchestration, higher-order workflows, durable execution, or remote agent hosting.
9. Validate with real providers, realistic prompts, and evaluation gates so the abstraction layer actually buys portability and reliability.

## Architecture

```mermaid
flowchart LR
  A["Task"] --> B{"Need agent threads, multi-agent orchestration, or remote agent hosting?"}
  B -->|Yes| C["Use Microsoft Agent Framework on top of `Microsoft.Extensions.AI.Abstractions`"]
  B -->|No| D{"Need provider-agnostic chat, embeddings, tools, typed output, or evaluation?"}
  D -->|Yes| E["Use `Microsoft.Extensions.AI`"]
  E --> F["Compose `IChatClient` / `IEmbeddingGenerator` in DI"]
  F --> G["Add caching, telemetry, tools, vector data, and evaluation deliberately"]
  D -->|No| H["Use plain provider SDKs or deterministic .NET code"]
```

## Core Knowledge

- `Microsoft.Extensions.AI.Abstractions` contains the core exchange contracts such as `IChatClient`, `IEmbeddingGenerator<TInput, TEmbedding>`, message/content types, and tool abstractions.
- `Microsoft.Extensions.AI` adds the higher-level application surface: middleware builders, automatic function invocation, caching, logging, and OpenTelemetry integration.
- Most apps and services should reference `Microsoft.Extensions.AI`; provider and connector libraries usually reference only the abstractions package.
- `IChatClient` centers on `GetResponseAsync` and `GetStreamingResponseAsync`. The returned `ChatResponse` or `ChatResponseUpdate` objects carry messages, tool-related content, metadata, and optional conversation identifiers.
- Local-model quickstarts still route through the same `IChatClient` abstraction. Ollama-backed clients are useful for low-cost prototyping, offline dev loops, and portability testing, but you still own chat history replay, latency, and model-quality tradeoffs.
- `ChatOptions` is the normal control plane for model ID, temperature, tools, `AdditionalProperties`, and provider-specific raw options.
- Tool calling is modeled with `AIFunction`, `AIFunctionFactory`, and `FunctionInvokingChatClient`. Ambient data can flow through closures, `AdditionalProperties`, `AIFunctionArguments.Context`, or DI.
- Tool calling can target local .NET methods, external APIs, or MCP-backed tools. The model requests calls; your app still owns execution, validation, and side-effect boundaries.
- Tool definitions consume request tokens. Keep tool descriptions short and register only the tools relevant for the current conversation or workflow.
- `FunctionInvokingChatClient` can handle the tool-invocation loop and parallel tool-call responses automatically when the provider/model supports that shape.
- `IEmbeddingGenerator` is the standard abstraction for semantic search, vector indexing, similarity, and cache-key generation. Pair it with `Microsoft.Extensions.VectorData.Abstractions` for vector store operations, and keep the embedding model, collection dimensions, and chunking/versioning story aligned so reindexing stays explicit.
- `IImageGenerator` is the experimental MEAI image surface. Treat `MEAI001` as an intentional opt-in, keep image generation separate from chat concerns, and compose logging/caching/hosting middleware around it the same way you would for `IChatClient`.
- `Microsoft.Extensions.DataIngestion` gives you the document-side RAG pipeline: `IngestionDocument`, document readers like MarkItDown/Markdig, document processors such as `ImageAlternativeTextEnricher`, chunkers, chunk processors, `VectorStoreWriter<T>`, and `IngestionPipeline<T>` for end-to-end composition.
- `IngestionPipeline<T>.ProcessAsync` is partial-success oriented. Handle `IAsyncEnumerable<IngestionResult>` deliberately instead of assuming one failed document should automatically crash the whole ingestion run.
- `Microsoft.Extensions.AI.Evaluation.*` gives you quality, NLP, safety, caching, and reporting layers for regression checks and CI gates.
- `dotnet/extensions` `v10.9.0` adds experimental `RoutingChatClient`/`SemanticRoutingChatClient` and `FailoverChatClient`/`OrderedFailoverChatClient` pipelines. Keep routing policy, fallback order, retry ownership, cost, and telemetry explicit; do not compose nested retry and failover layers without bounded attempts.
- Experimental failover switches clients only when the failed attempt has produced no output. Once streaming output has been emitted, surface the failure instead of replaying the response through another model and duplicating visible text or tool work. Test both failure-before-first-update and failure-after-first-update paths; keep fallback attempts bounded.
- The same release redesigns AI evaluation reports and refreshes their viewer. Treat report shape as a versioned CI artifact, and revalidate downstream parsers or publishing jobs before upgrading evaluation packages.
- The prior `v10.8.4` templates remove GitHub Models and require an explicit `--provider azureopenai`, `--provider ollama`, or `--provider openai`; update scaffolding scripts and provider-authentication tests instead of relying on the old default.
- The preceding `v10.8.0` release moved `Microsoft.Extensions.AI.OpenAI` to OpenAI 2.12.0, added speech-format auto-detection, and fixed `ImageGeneratingChatClient` content ordering. Keep multimodal and speech fixtures alongside the new approval/state tests.
- `AIFunctionNameAttribute`, `AIParameterNameAttribute`, and `ToolApprovalRequestContent.RequiresConfirmation` are new experimental `MEAI001` APIs. Opt in deliberately and keep approval decisions at the side-effect boundary.
- The August 2026 local official-docs snapshot mirrors the current 64-page `.NET AI` markdown tree, including the renamed tool-calling concept, MEDI/MEVD concepts, quickstart include fragments, and the dedicated vector-store section. Use `mcp` when the protocol itself becomes the design problem; stay here when you still mostly need app composition around `IChatClient` and friends.
- The current `.NET AI` ecosystem guidance separates direct MEAI composition, MCP interoperability, a prebuilt Copilot SDK harness, and Microsoft Agent Framework orchestration. Use `Microsoft Agent Framework` when you need autonomous orchestration, threads, workflows, hosting, or multi-agent collaboration instead of just model composition.

## Decision Cheatsheet

| If you need | Default choice | Why |
|---|---|---|
| App-level provider abstraction with middleware | `Microsoft.Extensions.AI` | Highest leverage for apps and services |
| A reusable provider or connector library | `Microsoft.Extensions.AI.Abstractions` | Keeps your package at the contract layer |
| Typed chat or UI streaming | `IChatClient` with `GetResponseAsync` / `GetStreamingResponseAsync` | Common request/response shape across providers |
| Tool calling from .NET methods | `AIFunction` + `FunctionInvokingChatClient` | Native function metadata and invocation pipeline |
| Typed structured output | `IChatClient.GetResponseAsync<T>` extensions | Keeps schema intent in code instead of prompt parsing |
| Vector search or RAG | `IEmbeddingGenerator` + `Microsoft.Extensions.VectorData.Abstractions` | Standardizes embeddings and store access |
| Local model prototyping | `IChatClient` with an Ollama-backed implementation | Keeps the app on the MEAI abstractions while you validate prompts or UX locally |
| Text-to-image or image-generation middleware | `IImageGenerator` | Use the dedicated image abstraction instead of overloading chat APIs |
| Evaluation and regression gates | `Microsoft.Extensions.AI.Evaluation.*` | Relevance, safety, task adherence, caching, reports |
| Agent threads or multi-step autonomous orchestration | `microsoft-agent-framework` | This is beyond plain provider abstraction |

## Common Failure Modes

- Referencing only `Microsoft.Extensions.AI.Abstractions` in an app and then rebuilding middleware, telemetry, or function invocation by hand.
- Treating `IChatClient` as if it already gives you durable agent threads, orchestration, or hosted-agent semantics.
- Mixing provider-specific assistants APIs with `IChatClient` as if they were the same runtime contract.
- Forgetting to distinguish stateless history replay from stateful `ConversationId` flows.
- Hiding important chat behavior in singleton service fields instead of explicit message history, options, or persistent storage.
- Adding tool calling without validating parameter binding, invalid input behavior, side effects, or DI-scoped dependencies.
- Building RAG without stable chunking, embedding-model/version tracking, or vector dimension discipline.
- Shipping AI features without evaluation baselines, safety checks, or telemetry for prompt/model drift.

## Deliver

- a justified package and abstraction choice: `Abstractions` only vs full `Microsoft.Extensions.AI`
- a concrete `IChatClient` / `IEmbeddingGenerator` composition strategy
- explicit tool-calling, options, state, caching, logging, and telemetry decisions
- vector-search, evaluation, or MCP integration guidance when the scenario needs it
- a clear escalation path to Agent Framework when the problem exceeds provider abstraction

## Validate

- the abstraction layer solves a real portability, testability, or composition problem
- provider registration and middleware order stay explicit in DI
- chat state management matches whether the provider is stateless or stateful
- structured output, tool invocation, and embedding flows are typed and observable
- vector store, embedding model, and chunking strategy are consistent
- evaluation or safety gates exist for important prompts and agent-like behaviors
- agentic requirements are not being under-modeled as a simple `IChatClient` integration

When exact wording, edge-case API behavior, or less-common examples matter, check the local official docs snapshot before relying on summaries.

## References

- [official-docs-index.md](references/official-docs-index.md) - Slim local snapshot map with direct links to every mirrored `.NET AI` docs page plus API-reference pointers
- [patterns.md](references/patterns.md) - Package choice, `IChatClient`, embeddings, DI pipelines, tool-calling, and Agent Framework escalation guidance
- [examples.md](references/examples.md) - Quickstart-to-task map covering chat, structured output, function calling, vector search, local models, MCP, and assistants
- [evaluation.md](references/evaluation.md) - Quality, NLP, safety, caching, reporting, and CI-oriented evaluation guidance

© managedcode · MIT · managedcode/dotnet-skills

Similar skills

All skills for AI and ML →
microsoft-extensions-configuration Aaronontheweb microsoft-extensions-configuration — a skill for Claude and other AI editors that guides binding appsettings.json to strongly-typed classes and validating configuration at startup. CX Codex +1 621installs microsoft-azure-webjobs-extensions-authentication-events-dotnet Microsoft microsoft-azure-webjobs-extensions-authentication-events-dotnet is a Claude Code skill for building .NET Azure Functions that handle Microsoft Entra authentication events. CX Codex +1 1installs rag-architect https://github.com/Jeffallan rag-architect designs production RAG systems: chunks documents, generates embeddings, configures vector stores, builds hybrid search pipelines, and evaluates retrieval quality. CX Codex +1 5,100installs langchain4j-rag-implementation-patterns giuseppe-trisciuoglio langchain4j-rag-implementation-patterns — a coding-agent skill that scaffolds LangChain4j RAG pipelines: document ingestion, embedding stores, vector search and knowledge-base answers. CX Codex +1 3,900installs AgentDB Vector Search ruvnet AgentDB Vector Search — a skill for fast semantic document retrieval using AgentDB's vector database. Use it for RAG systems, search engines, and knowledge bases where you need relevant results from large text collections. CX Codex +1 1,100installs upstash-vector-db-skills gocallum The upstash-vector-db-skills skill walks the agent through Upstash Vector: index setup, document upserts, semantic search and namespaces. Useful when you build vector search on Vercel. CX Codex +1 369installs

From the same set38

The whole set →
signalr Guides the implementation and review of SignalR hubs, streaming, groups, and reconnection logic in ASP.NET Core. CX Codex +1 ★ 521repo stars code-testing-implementer The code-testing-implementer subagent implements one test-plan phase, writes the test files, and checks whether they build and pass. Use it to carry out a phase or run a build-test-fix cycle. CX Codex ★ 521repo stars mcp mcp guides building and consuming Model Context Protocol servers and clients in C# on .NET 8+, covering stdio, Streamable HTTP, and tools. CX Codex +1 ★ 521repo stars unit-test-generation The unit-test-generation prompt guides an agent through writing focused unit tests that check behavior and catch plausible bugs. Use it when adding tests to existing code or starting a new test suite. CX Codex +1 ★ 521repo stars wpf Rules and patterns for WPF apps: bindings, commands, threading, styles. Fits UI work, MVVM refactoring, or moving a project from .NET Framework to .NET. CX Codex +1 ★ 521repo stars worker-services worker-services guides the agent in building long-running .NET background services with BackgroundService, PeriodicTimer, and proper dependency scoping. CX Codex +1 ★ 521repo stars aspire This skill guides work on Aspire AppHosts for distributed C# and TypeScript applications. Use it to wire services, configure integrations, or test an Aspire topology. CX Codex +1 ★ 521repo stars improve-skill-quality improve-skill-quality — a skill for the dotnet/skills repository that traces regressions, timeouts, activation failures, and inconclusive evaluation results back to their source. CX Codex +1 ★ 521repo stars

Project sponsors

Companies that keep this analytics open