### Guide Model Behavior with System Instructions

Source: https://github.com/mscraftsman/generative-ai/blob/main/docs/docs/text-generation.md

Provide system instructions to guide the model's responses. This example sets the model to respond as a friendly pirate.

```csharp
using Mscc.GenerativeAI;

var systemInstruction = new Content("You are a friendly pirate. Speak like one.");
var prompt = "Good morning! How are you?";
var googleAI = new GoogleAI(apiKey: "your API key");
var model = googleAI.GenerativeModel(model: Model.Gemini25Flash, 
    systemInstruction: systemInstruction);
var request = new GenerateContentRequest(prompt);

var response = await model.GenerateContent(request);
Console.WriteLine(response.Text);
```

--------------------------------

### Run the Application

Source: https://github.com/mscraftsman/generative-ai/blob/main/samples/Console.Gemini.FunctionCalling/README.md

Execute the .NET application to start the interactive console experience.

```bash
dotnet run
```

--------------------------------

### User Input Example

Source: https://github.com/mscraftsman/generative-ai/blob/main/samples/Console.Gemini.FunctionCalling/README.md

Provide input to the application to test function calling. Questions that can be answered by the defined tools will trigger tool calls.

```text
User > What is my name?
```

--------------------------------

### Install Mscc.GenerativeAI.Web via NuGet Package Manager

Source: https://github.com/mscraftsman/generative-ai/blob/main/src/Mscc.GenerativeAI.Web/README.md

Install the Mscc.GenerativeAI.Web NuGet package using the Package Manager Console in Visual Studio.

```powershell
PM> Install-Package Mscc.GenerativeAI.Web
```

--------------------------------

### Make first Gemini API request using C#

Source: https://github.com/mscraftsman/generative-ai/blob/main/docs/index.md

This example demonstrates how to initialize the Gemini client and use the GenerateContent method to send a request to the Gemini API. It assumes the GEMINI_API_KEY environment variable is set.

```csharp
using Mscc.GenerativeAI;

# The client gets the API key from the environment variable `GEMINI_API_KEY`.
var googleAI = new GoogleAI();  // or pass in apiKey parameter
var model = googleAI.GenerativeModel(model: Model.Gemini25Flash);

var response = await model.GenerateContent("Explain how AI works in a few words");
Console.WriteLine(response.Text);
```

--------------------------------

### Add Mscc.GenerativeAI.Web Package

Source: https://github.com/mscraftsman/generative-ai/blob/main/src/Mscc.GenerativeAI.Web/README.md

Install the Mscc.GenerativeAI.Web NuGet package using the dotnet CLI.

```text
> dotnet add package Mscc.GenerativeAI.Web
```

--------------------------------

### Generate Content from Text and Image Input

Source: https://github.com/mscraftsman/generative-ai/blob/main/README.md

This example shows how to process an image along with a text prompt. It requires the `Mscc.GenerativeAI` namespace and a valid API key.

```csharp
using Mscc.GenerativeAI;

var apiKey = "your_api_key";
var prompt = "Parse the time and city from the airport board shown in this image into a list, in Markdown";
var googleAI = new GoogleAI(apiKey: apiKey);
var model = googleAI.GenerativeModel(model: Model.Gemini25Flash);
var request = new GenerateContentRequest(prompt);
await request.AddMedia("https://raw.githubusercontent.com/mscraftsman/generative-ai/refs/heads/main/tests/Mscc.GenerativeAI/payload/timetable.png");

var response = await model.GenerateContent(request);
Console.WriteLine(response.Text);
```

--------------------------------

### Configure Generative AI with Configuration Section Path

Source: https://github.com/mscraftsman/generative-ai/blob/main/src/Mscc.GenerativeAI.Web/README.md

This approach simplifies configuration by directly providing the name of the configuration section as a string. It's useful for straightforward setups where the section name is known.

```csharp
var builder = WebApplication.CreateBuilder(args);

builder.Services.AddGenerativeAI("Gemini");

var app = builder.Build();
// ...
```

--------------------------------

### Upload and Use Large Files for Content Generation

Source: https://github.com/mscraftsman/generative-ai/blob/main/README.md

This example shows how to upload large media files using the File API and then reference them in a prompt for content generation. Ensure the `Mscc.GenerativeAI` namespace and necessary file paths are correctly set.

```csharp
using Mscc.GenerativeAI;

var apiKey = "your_api_key";
var prompt = "Make a short story from the media resources. The media resources are:";
IGenerativeAI genAi = new GoogleAI(apiKey);
var model = genAi.GenerativeModel(Model.Gemini25Flash);

// Upload your large image(s).
// Instead of discarding you could also use the response and access `response.Text`.
var filePath = Path.Combine(Environment.CurrentDirectory, "verylarge.png");
var displayName = "My very large image";
_ = await model.UploadMedia(filePath, displayName);

// Create the prompt with references to File API resources.
var request = new GenerateContentRequest(prompt);
var files = await model.ListFiles();
foreach (var file in files.Where(x => x.MimeType.StartsWith("image/", StringComparison.OrdinalIgnoreCase)))
{
    Console.WriteLine($"File: {file.Name}");
    request.AddMedia(file);
}
var response = await model.GenerateContent(request);
Console.WriteLine(response.Text);
```

--------------------------------

### Navigate to Sample Directory

Source: https://github.com/mscraftsman/generative-ai/blob/main/samples/Console.Gemini.FunctionCalling/README.md

Change your current directory to the sample's location within the project structure.

```bash
cd samples/Console.Gemini.FunctionCalling
```

--------------------------------

### Pack the main SDK for .NET

Source: https://github.com/mscraftsman/generative-ai/blob/main/AGENTS.md

Create NuGet packages for the main SDK targeting .NET. The output packages will be placed in the 'output/' directory.

```bash
dotnet pack -c Release ./src/Mscc.GenerativeAI/Mscc.GenerativeAI.csproj -o output/
```

--------------------------------

### Install Mscc.GenerativeAI via .NET CLI

Source: https://github.com/mscraftsman/generative-ai/blob/main/README.md

Use this command to add the Mscc.GenerativeAI package to your .NET project from the command line.

```text
> dotnet add package Mscc.GenerativeAI
```

--------------------------------

### Create new ASP.NET Core Web Project

Source: https://github.com/mscraftsman/generative-ai/blob/main/src/Mscc.GenerativeAI.Web/README.md

Use the dotnet CLI to create a new minimal API project.

```text
> dotnet new web -o Web.Minimal.Api
```

--------------------------------

### Build the entire solution

Source: https://github.com/mscraftsman/generative-ai/blob/main/AGENTS.md

Use this command to build the complete .NET solution. It compiles all projects within the GenerativeAI.sln file in Release configuration.

```bash
dotnet build ./GenerativeAI.sln -c Release
```

--------------------------------

### Add Mscc.GenerativeAI package using dotnet CLI

Source: https://github.com/mscraftsman/generative-ai/blob/main/docs/index.md

Install the Mscc.GenerativeAI NuGet package using the dotnet command-line tool in your .NET project folder.

```text
dotnet add package Mscc.GenerativeAI
```

--------------------------------

### Configure Generative AI with Action<TOptions>

Source: https://github.com/mscraftsman/generative-ai/blob/main/src/Mscc.GenerativeAI.Web/README.md

Configure the service by providing an Action delegate that accepts a TOptions object. This allows for programmatic setting of options, including sensitive values like API keys, directly within your code.

```csharp
var builder = WebApplication.CreateBuilder(args);

builder.Services.AddGenerativeAI(options =>
{
    // User defined option values
    options.ProjectId = string.Empty;
    options.Model = GenerativeAI.Types.Model.GeminiProVision;
    options.Credentials.ApiKey = "YOUR_API_KEY";
});

var app = builder.Build();
// ...
```

--------------------------------

### Run all tests using Makefile

Source: https://github.com/mscraftsman/generative-ai/blob/main/AGENTS.md

Conveniently run all test projects within the solution using the provided Makefile. This is the recommended approach for comprehensive testing.

```bash
make test
```

--------------------------------

### Add Mscc.GenerativeAI package using Package Manager Console

Source: https://github.com/mscraftsman/generative-ai/blob/main/docs/index.md

Install the Mscc.GenerativeAI NuGet package using the Package Manager Console in Visual Studio.

```powershell
Install-Package Mscc.GenerativeAI
```

--------------------------------

### Pack the Google Cloud Client Library SDK

Source: https://github.com/mscraftsman/generative-ai/blob/main/AGENTS.md

Create NuGet packages for the SDK version that utilizes the Google Cloud Client Library. The generated packages will be saved in the 'output/' directory.

```bash
dotnet pack -c Release ./src/Mscc.GenerativeAI.Google/Mscc.GenerativeAI.Google.csproj -o output/
```

--------------------------------

### Generate Content with Multimodal Input (Image)

Source: https://github.com/mscraftsman/generative-ai/blob/main/docs/docs/text-generation.md

Combine text prompts with media files, such as images, for richer context. This example parses information from an airport board image.

```csharp
using Mscc.GenerativeAI;

var prompt = "Parse the time and city from the airport board shown in this image into a list, in Markdown";
var googleAI = new GoogleAI();
var model = googleAI.GenerativeModel(model: Model.Gemini25Flash);
var request = new GenerateContentRequest(prompt);
await request.AddMedia("https://raw.githubusercontent.com/mscraftsman/generative-ai/refs/heads/main/tests/Mscc.GenerativeAI/payload/timetable.png");

var response = await model.GenerateContent(request);
Console.WriteLine(response.Text);
```

--------------------------------

### Streaming Content Generation with Tools

Source: https://github.com/mscraftsman/generative-ai/blob/main/src/Mscc.GenerativeAI.Microsoft/README.md

Configure the IChatClient for streaming responses and enable function invocation and OpenTelemetry. This example demonstrates using a custom tool to retrieve age information.

```csharp
using Microsoft.Extensions.AI;
using Mscc.GenerativeAI.Microsoft;
using System.ComponentModel

// assuming credentials are set up in environment variables as instructed above.
IChatClient chatClient = new GeminiChatClient()
    .AsIChatClient("gemini-2.5-flash")
    .AsBuilder()
    .UseFunctionInvocation()
    .UseOpenTelemetry()
    .Build();

ChatOptions options = new() 
{
    Tools = [AIFunctionFactory.Create(([Description("The name of the person whose age is to be retrieved")] string personName) => personName switch
    {
        "Alice" => 30,
        "Bob" => 25,
        _ => 35
    }, "get_person_age", "Gets the age of the specified person");
};

await foreach (var update in chatClient.GetStreamingResponseAsync("How much older is Alice than Bob?", options))
{
    Console.Write(update);
}
```

--------------------------------

### Initiate and Manage Chat Conversations

Source: https://github.com/mscraftsman/generative-ai/blob/main/README.md

This snippet demonstrates how to start a multi-turn chat conversation. It uses the default Gemini Pro model and allows for managing chat history, including rewinding.

```csharp
using Mscc.GenerativeAI;

var apiKey = "your_api_key";
var googleAI = new GoogleAI(apiKey);
var model = googleAI.GenerativeModel();    // using default model: gemini-2.5-pro
var chat = model.StartChat();   // optionally pass a previous history in the constructor.

// Instead of discarding you could also use the response and access `response.Text`.
_ = await chat.SendMessage("Hello, fancy brainstorming about IT?");
_ = await chat.SendMessage("In one sentence, explain how a computer works to a young child.");
_ = await chat.SendMessage("Okay, how about a more detailed explanation to a high schooler?");
_ = await chat.SendMessage("Lastly, give a thorough definition for a CS graduate.");

// A chat session keeps every response in its history.
chat.History.ForEach(c => Console.WriteLine($"{c.Role}: {c.Text}"));

// Last request/response pair can be removed from the history.
var latest = chat.Rewind();
Console.WriteLine($"{latest.Sent} - {latest.Received}");
```