Skip to content

Latest commit

 

History

History
105 lines (77 loc) · 3.36 KB

develop.md

File metadata and controls

105 lines (77 loc) · 3.36 KB
Raw

Local Development

Required Tools

Azure Functions Core Tools

  • Install

  • Create a folder (c:\temp\sk) and open a command prompt and navigate to that folder cd c:\temp\sk

  • Run the following command to create a new console project

dotnet new console

Instantiate the kernel and run it

dotnet add package Microsoft.SemanticKernel --prerelease

using Microsoft.SemanticKernel;

// Set Simple kernel instance
IKernel kernel_1 = KernelBuilder.Create();

Add built in skills

using Microsoft.SemanticKernel.Skills.Core;

// Load native skill
var text = kernel.ImportSkill(new TextSkill());

SKContext result = await kernel.RunAsync("    i n f i n i t e     s p a c e     ",
    text["TrimStart"],
    text["TrimEnd"],
    text["Uppercase"]);

Console.WriteLine(result);

Overall Create kernel outline to add add model (brains) add plugin - native and llm prompts

Pre-requisites Create Azure Open AI endpoint and chatgpt deployment

Create project dotnet new console add following packages Microsoft.Extensions.Configuration Microsoft.Extensions.Configuration.Abstractions Microsoft.Extensions.Configuration.Binder Microsoft.Extensions.Configuration.EnvironmentVariables Microsoft.Extensions.Configuration.Json Microsoft.Extensions.Configuration.UserSecrets Microsoft.Extensions.DependencyInjection System.Linq.Async

Review PluginConfiguration.cs

Secret Manager to avoid the risk of leaking secrets into the repository, branches and pull requests. You can also use environment variables if you prefer.

To set your secrets with Secret Manager (inside project folder):

dotnet user-secrets init

then add azure open ai secrets dotnet user-secrets set "AzureOpenAI:DeploymentName" "..." dotnet user-secrets set "AzureOpenAI:Endpoint" "https://... .openai.azure.com/" dotnet user-secrets set "AzureOpenAI:ApiKey" "..."

dotnet add package Microsoft.SemanticKernel --prerelease

review chatexample.cs which shows how to use built-in fileio and summarize skill along with chatgpt

use writerskill to rewrite Lincoln's Gettysburg address in modern English using chatgpt Take the output and summarize it using the built-in summarization skill Shows planner with sequential

Prep work setup up class that connects to azure open ai endpoint and initializes kernel

We will show how to use built-in skills, how to use chat-gpt to read links and summarize it, and how to use the planner to create a sequential pipeline of skills.

  1. Summarize text block with built-in text skills dotnet new console dotnet add package Microsoft.SemanticKernel --prerelease

  2. Summarize text block with plugin

  3. Use chat-gpt to read a link and summarize it with built-in skills

  • orchestrate with planner
  1. Use chat-gpt to read multiple links from memory and summarize them with built-in skills
  1. Use chat-gpt to read multiple links from memory and summarize them with built-in skills and recommend additional reading sources
  • Show how to use planner to create a sequential pipeline of skills