Conversation
There was a problem hiding this comment.
The meat of the changes is in this file.
There was a problem hiding this comment.
Pull request overview
This PR removes references to Semantic Kernel from AI documentation and replaces them with Microsoft Agent Framework where applicable, contributing to issue #51515. The changes align with the goal of deprecating Semantic Kernel in favor of Agent Framework for new .NET AI applications.
Changes:
- Removed Semantic Kernel include files and references from quickstart documentation
- Added new code snippet demonstrating multi-turn chat using Agent Framework
- Updated conceptual documentation to replace Semantic Kernel references with Agent Framework
- Removed Semantic Kernel connector documentation for vector databases
Reviewed changes
Copilot reviewed 19 out of 19 changed files in this pull request and generated 3 comments.
Show a summary per file
| File | Description |
|---|---|
| docs/ai/snippets/prompt-engineering/multi-turn-chat.csproj | New project file for Agent Framework multi-turn chat example |
| docs/ai/snippets/prompt-engineering/multi-turn-chat.cs | New code snippet demonstrating Agent Framework usage |
| docs/ai/quickstarts/prompt-model.md | Removed Semantic Kernel include reference |
| docs/ai/quickstarts/includes/semantic-kernel.md | Deleted Semantic Kernel informational include file |
| docs/ai/quickstarts/generate-images.md | Removed Semantic Kernel include reference |
| docs/ai/quickstarts/chat-local-model.md | Removed Semantic Kernel SDK reference from description |
| docs/ai/quickstarts/build-vector-search-app.md | Removed Semantic Kernel collaboration mention |
| docs/ai/quickstarts/build-chat-app.md | Removed Semantic Kernel include reference |
| docs/ai/overview.md | Removed Semantic Kernel from provider compatibility statement |
| docs/ai/index.yml | Reordered links and updated training references |
| docs/ai/includes/sk-connectors.md | Deleted Semantic Kernel vector database connectors documentation |
| docs/ai/get-started-app-chat-template.md | Replaced Semantic Kernel with Agent Framework in architecture description |
| docs/ai/dotnet-ai-ecosystem.md | Updated Semantic Kernel section to recommend Agent Framework for new apps |
| docs/ai/conceptual/vector-databases.md | Removed Semantic Kernel vector database solutions section |
| docs/ai/conceptual/understanding-tokens.md | Removed Semantic Kernel vector store connectors link |
| docs/ai/conceptual/understanding-openai-functions.md | Removed Semantic Kernel function calling section |
| docs/ai/conceptual/prompt-engineering-dotnet.md | Replaced Semantic Kernel references with Agent Framework |
| docs/ai/conceptual/embeddings.md | Updated to reference Agent Framework instead of Semantic Kernel |
| docs/ai/azure-ai-services-authentication.md | Updated authentication example to use Azure OpenAI instead of Semantic Kernel |
| - [How GenAI and LLMs work](how-genai-and-llms-work.md) | ||
| - [Retrieval-augmented generation](rag.md) | ||
| - [Training: Develop AI agents with Azure OpenAI and Semantic Kernel](/training/paths/develop-ai-agents-azure-open-ai-semantic-kernel-sdk/) | ||
| - [Training: Develop an AI agent with Microsoft Agent Framework](/training/modules/develop-ai-agent-with-semantic-kernel/) |
There was a problem hiding this comment.
The link text says "Training: Develop an AI agent with Microsoft Agent Framework" but the URL points to /training/modules/develop-ai-agent-with-semantic-kernel/, which appears to be Semantic Kernel training content. This is inconsistent with the PR's goal of removing Semantic Kernel references. The URL should be updated to point to Agent Framework training content if it exists, or the link should be removed if Agent Framework training is not yet available.
| - [Training: Develop an AI agent with Microsoft Agent Framework](/training/modules/develop-ai-agent-with-semantic-kernel/) |
| - LLMs have token input limits that constrain the amount of text you can fit in a prompt. Use [embeddings](embeddings.md) and [vector database solutions](vector-databases.md) to reduce the number of tokens you need to represent a given piece of text. | ||
| - LLMs aren't trained on your data unless you train them yourself, which can be costly and time-consuming. Use [retrieval augmented generation (RAG)](rag.md) to make your data available to an LLM without training it. | ||
| - Creates an AI agent using a chat client configured for a specific model endpoint. | ||
| - Configures the agent with system instructions using `WithInstructions`. |
There was a problem hiding this comment.
The documentation states the code "Configures the agent with system instructions using WithInstructions", but the referenced code sample (multi-turn-chat.cs) doesn't actually call WithInstructions or demonstrate configuring system instructions. The code should either be updated to include system instructions, or this bullet point should be removed or modified to accurately reflect what the code demonstrates.
| - text: Develop an AI agent with Microsoft Agent Framework | ||
| url: /training/modules/develop-ai-agent-with-semantic-kernel/ |
There was a problem hiding this comment.
The link text says "Develop an AI agent with Microsoft Agent Framework" but the URL points to /training/modules/develop-ai-agent-with-semantic-kernel/, which appears to be Semantic Kernel training content. This is inconsistent with the PR's goal of removing Semantic Kernel references. The URL should be updated to point to Agent Framework training content if it exists, or the link should be removed if Agent Framework training is not yet available.
| - text: Develop an AI agent with Microsoft Agent Framework | |
| url: /training/modules/develop-ai-agent-with-semantic-kernel/ |
1 participant
Contributes to #51515.
Internal previews
Toggle expand/collapse