semantic-kernel

Name: semantic-kernel
Author: managedcode

461stars35forks

How to install semantic-kernel?

managedcode/dotnet-skills/semantic-kernel

$npx -y skills add managedcode/dotnet-skills --skill semantic-kernel

Installs into the current project.

Summary

Build AI-enabled .NET applications with Semantic Kernel using services, plugins, prompts, and function-calling patterns that remain testable and maintainable. USE FOR: adding AI-driven prompts, plugins, or orchestration to a .NET app; reviewing kernel construction, service regist

Files · 1

View on GitHub

SKILL.md

1	# Semantic Kernel for .NET
2
3	## Trigger On
4
5	- adding AI-driven prompts, plugins, or orchestration to a .NET app
6	- reviewing kernel construction, service registration, or plugin usage
7	- building function-calling patterns with LLMs
8	- migrating older Semantic Kernel code to current APIs
9
10	## Documentation
11
12	- [Semantic Kernel Overview](https://learn.microsoft.com/en-us/semantic-kernel/overview/)
13	- [Plugins and Functions](https://learn.microsoft.com/en-us/semantic-kernel/concepts/plugins/)
14	- [Agent Functions](https://learn.microsoft.com/en-us/semantic-kernel/frameworks/agent/agent-functions)
15	- [GitHub Repository](https://github.com/microsoft/semantic-kernel)
16	- [Microsoft Agent Framework](https://learn.microsoft.com/en-us/semantic-kernel/frameworks/agent/)
17
18	### References
19
20	- [patterns.md](references/patterns.md) - Plugin patterns, function calling patterns, multi-agent patterns, prompt templates, and RAG patterns
21	- [anti-patterns.md](references/anti-patterns.md) - Common Semantic Kernel mistakes and how to avoid them
22
23	## Core Concepts
24
25	\| Concept \| Description \|
26	\|---------\|-------------\|
27	\| Kernel \| Central orchestrator for AI services and plugins \|
28	\| Plugin \| Collection of functions exposed to the LLM \|
29	\| Function \| Native C# method or prompt template \|
30	\| Chat Completion \| LLM service for generating responses \|
31	\| Memory \| Vector storage for semantic search \|
32
33	## Workflow
34
35	1. Build the Kernel with required services
36	2. Create Plugins with well-described functions
37	3. Configure Function Calling for automatic tool use
38	4. Handle Responses and manage conversation state
39	5. Test and Observe AI behavior with logging
40	6. For Semantic Kernel `dotnet-1.78.0` and later, keep OpenAPI plugin server URL validation enabled, do not re-enable automatic redirects on the default `HttpPlugin` or `WebFileDownloadPlugin` clients without an explicit trusted-host policy, and use the current Microsoft Agent Framework-compatible migration samples when moving SK agent code to Agent Framework.
41	7. Re-test custom file, document, and web plugin paths after upgrading to `1.78.0`; the release hardens path validation and updates vulnerable transitive dependencies, so local workarounds that weakened validation should be removed rather than carried forward.
42
43	## Kernel Setup
44
45	### Basic Configuration
46	```csharp
47	var builder = Kernel.CreateBuilder();
48
49	builder.AddAzureOpenAIChatCompletion(
50	deploymentName: "gpt-4",
51	endpoint: config["AzureOpenAI:Endpoint"]!,
52	apiKey: config["AzureOpenAI:ApiKey"]!);
53
54	// Or OpenAI
55	builder.AddOpenAIChatCompletion(
56	modelId: "gpt-4",
57	apiKey: config["OpenAI:ApiKey"]!);
58
59	var kernel = builder.Build();
60	```
61
62	### With Dependency Injection
63	```csharp
64	builder.Services.AddKernel()
65	.AddAzureOpenAIChatCompletion(
66	deploymentName: "gpt-4",
67	endpoint: config["AzureOpenAI:Endpoint"]!,
68	apiKey: config["AzureOpenAI:ApiKey"]!);
69
70	// Register plugins
71	builder.Services.AddSingleton<WeatherPlugin>();
72	builder.Services.AddSingleton<OrderPlugin>();
73
74	// In your service
75	public class AiService(Kernel kernel)
76	{
77	public async Task<string> ChatAsync(string message)
78	{
79	var response = await kernel.InvokePromptAsync(message);
80	return response.ToString();
81	}
82	}
83	```
84
85	## Plugin Patterns
86
87	### Creating a Plugin
88	```csharp
89	public class WeatherPlugin
90	{
91	[KernelFunction]
92	[Description("Gets the current weather for a specified city")]
93	public async Task<string> GetWeather(
94	[Description("The city name, e.g., 'Seattle'")] string city,
95	[Description("Temperature unit: 'celsius' or 'fahrenheit'")] string unit = "celsius")
96	{
97	// Call actual weather API
98	var weather = await _weatherService.GetCurrentAsync(city);
99	return $"Weather in {city}: {weather.Temperature}° {unit}, {weather.Condition}";
100	}
101
102	[KernelFunction]
103	[Description("Gets the weather forecast for the next N days")]
104	public async Task<string> GetForecast(
105	[Description("The city name")] string city,
106	[Description("Number of days (1-7)")] int days = 3)
107	{
108	var forecast = await _weatherService.GetForecastAsync(city, days);
109	return FormatForecast(forecast);
110	}
111	}
112	```
113
114	### Plugin Best Practices
115
116	\| Practice \| Why It Matters \|
117	\|----------\|----------------\|
118	\| Clear `[De

People who liked this also viewed

6 picks

Type

semantic-kernel

461stars35forks

View on GitHub

How to install semantic-kernel?

managedcode/dotnet-skills/semantic-kernel

$npx -y skills add managedcode/dotnet-skills --skill semantic-kernel

Installs into the current project.

Summary

Files · 1

View on GitHub

SKILL.md

1	# Semantic Kernel for .NET
2
3	## Trigger On
4
5	- adding AI-driven prompts, plugins, or orchestration to a .NET app
6	- reviewing kernel construction, service registration, or plugin usage
7	- building function-calling patterns with LLMs
8	- migrating older Semantic Kernel code to current APIs
9
10	## Documentation
11
12	- [Semantic Kernel Overview](https://learn.microsoft.com/en-us/semantic-kernel/overview/)
13	- [Plugins and Functions](https://learn.microsoft.com/en-us/semantic-kernel/concepts/plugins/)
14	- [Agent Functions](https://learn.microsoft.com/en-us/semantic-kernel/frameworks/agent/agent-functions)
15	- [GitHub Repository](https://github.com/microsoft/semantic-kernel)
16	- [Microsoft Agent Framework](https://learn.microsoft.com/en-us/semantic-kernel/frameworks/agent/)
17
18	### References
19
20	- [patterns.md](references/patterns.md) - Plugin patterns, function calling patterns, multi-agent patterns, prompt templates, and RAG patterns
21	- [anti-patterns.md](references/anti-patterns.md) - Common Semantic Kernel mistakes and how to avoid them
22
23	## Core Concepts
24
25	\| Concept \| Description \|
26	\|---------\|-------------\|
27	\| Kernel \| Central orchestrator for AI services and plugins \|
28	\| Plugin \| Collection of functions exposed to the LLM \|
29	\| Function \| Native C# method or prompt template \|
30	\| Chat Completion \| LLM service for generating responses \|
31	\| Memory \| Vector storage for semantic search \|
32
33	## Workflow
34
35	1. Build the Kernel with required services
36	2. Create Plugins with well-described functions
37	3. Configure Function Calling for automatic tool use
38	4. Handle Responses and manage conversation state
39	5. Test and Observe AI behavior with logging
40	6. For Semantic Kernel `dotnet-1.78.0` and later, keep OpenAPI plugin server URL validation enabled, do not re-enable automatic redirects on the default `HttpPlugin` or `WebFileDownloadPlugin` clients without an explicit trusted-host policy, and use the current Microsoft Agent Framework-compatible migration samples when moving SK agent code to Agent Framework.
41	7. Re-test custom file, document, and web plugin paths after upgrading to `1.78.0`; the release hardens path validation and updates vulnerable transitive dependencies, so local workarounds that weakened validation should be removed rather than carried forward.
42
43	## Kernel Setup
44
45	### Basic Configuration
46	```csharp
47	var builder = Kernel.CreateBuilder();
48
49	builder.AddAzureOpenAIChatCompletion(
50	deploymentName: "gpt-4",
51	endpoint: config["AzureOpenAI:Endpoint"]!,
52	apiKey: config["AzureOpenAI:ApiKey"]!);
53
54	// Or OpenAI
55	builder.AddOpenAIChatCompletion(
56	modelId: "gpt-4",
57	apiKey: config["OpenAI:ApiKey"]!);
58
59	var kernel = builder.Build();
60	```
61
62	### With Dependency Injection
63	```csharp
64	builder.Services.AddKernel()
65	.AddAzureOpenAIChatCompletion(
66	deploymentName: "gpt-4",
67	endpoint: config["AzureOpenAI:Endpoint"]!,
68	apiKey: config["AzureOpenAI:ApiKey"]!);
69
70	// Register plugins
71	builder.Services.AddSingleton<WeatherPlugin>();
72	builder.Services.AddSingleton<OrderPlugin>();
73
74	// In your service
75	public class AiService(Kernel kernel)
76	{
77	public async Task<string> ChatAsync(string message)
78	{
79	var response = await kernel.InvokePromptAsync(message);
80	return response.ToString();
81	}
82	}
83	```
84
85	## Plugin Patterns
86
87	### Creating a Plugin
88	```csharp
89	public class WeatherPlugin
90	{
91	[KernelFunction]
92	[Description("Gets the current weather for a specified city")]
93	public async Task<string> GetWeather(
94	[Description("The city name, e.g., 'Seattle'")] string city,
95	[Description("Temperature unit: 'celsius' or 'fahrenheit'")] string unit = "celsius")
96	{
97	// Call actual weather API
98	var weather = await _weatherService.GetCurrentAsync(city);
99	return $"Weather in {city}: {weather.Temperature}° {unit}, {weather.Condition}";
100	}
101
102	[KernelFunction]
103	[Description("Gets the weather forecast for the next N days")]
104	public async Task<string> GetForecast(
105	[Description("The city name")] string city,
106	[Description("Number of days (1-7)")] int days = 3)
107	{
108	var forecast = await _weatherService.GetForecastAsync(city, days);
109	return FormatForecast(forecast);
110	}
111	}
112	```
113
114	### Plugin Best Practices
115
116	\| Practice \| Why It Matters \|
117	\|----------\|----------------\|
118	\| Clear `[De

People who liked this also viewed

6 picks

Type