Semantic Kernel

As explained in AI Agents in Action

Semantic Kernel is an open-source Microsoft software development kit for building AI applications that orchestrate language-model capabilities, plugins, and related services.

Semantic Kernel (SK) is Microsoft’s open-source SDK for building AI applications and agents. It orchestrates plugins, functions, language-model services, and chat interfaces so agents can act beyond conversation. section 5.3 section 5.8

Use SK to organize several kinds of agent capability:

• Define semantic functions that encapsulate prompt templates and execute them against a language model. section 5.3.1
• Wrap executable code and APIs as native functions, then combine them with semantic functions in layered execution stages. section 5.8
• Manage multiple plugins through a chat interface or agent, while consuming or publishing compatible OpenAI plugin definitions. section 5.3

Connect a chat service

Install SK in a Python workspace, create a kernel, load OpenAI or Azure OpenAI settings, register a chat-completion service, and execute a prompt as a semantic function. The same approach can expose an API through a GPT interface and make it available to chat or agent users. section 5.3.1 section 5.5

Figure 5.5 How the Semantic Kernel integrates as a plugin and can also consume plugins

Trace how SK can operate as a plugin, consume other plugins, and coordinate them behind a chat or agent interface. section 5.3

How AI Agents in Action covers it

How deep it goes

Demonstrates installation, service registration, semantic-function execution, plugin composition, and GPT-interface construction, while focusing on actions rather than memory or planning.

Examples

Uses OpenAI or Azure OpenAI chat completion, TMDB API integration, temperature conversion, weather, creative writing, meal planning, and news plugins.

What it assumes

Assumes a configured Python workspace and service credentials, while teaching semantic functions, native functions, and plugins from their basic roles.

What you can do after

Build agent actions by installing SK, registering language-model services, creating functions, combining plugins, and exposing APIs through chat interfaces.

How it connects

Connects OpenAI function calling, plugins, semantic functions, native functions, GPT interfaces, chat interfaces, memory, and planning.

liveBookmanning.com

Read AI Agents in Action on liveBook Get the book at manning.com

Also explained in

• 

  liveBookmanning.com
  Generative AI in ActionSK is an AI application framework that provides abstractions for integrating language models with prompts, functions, and external services.
• 

  liveBookmanning.com
  Building Distributed Applications that WorkSemantic Kernel is a collection of .NET libraries that provides interfaces for integrating large language models with application code.

Compare how 3 books explain Semantic Kernel →

Related concepts

• large language modeluses · A large language model is a machine-learning model, typically neural, trained on extensive text data to process and generate language.
• Ollamauses · Ollama is software for downloading, running, and serving language models locally.
• pluginuses · A plugin is an optional software component that extends or modifies a host application’s capabilities through an extension or integration mechanism.
• client librarykind of · A client library is reusable software that provides programmatic interfaces for an application to interact with a service, platform, or system.
• native functionconfigured by · A native function is a conventionally programmed function that can be invoked by an orchestration framework or language-model application.

This explainer was generated by Manning's knowledge graph from the text of AI Agents in Action. The section links point to the parts of the book it draws on.