### Traditional RAG API Request Example

Source: https://github.com/akshaykokane/implementing-agentic-rag-with-semantic-kernel-and-azure-ai-search/blob/main/README.md

An example of a POST request to the /rag/traditional API endpoint, demonstrating how to send a query for the Traditional RAG service.

```json
POST /rag/traditional
{
  "query": "Find best hotels in Sedona"
}
```

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

### Agentic RAG API Request Example

Source: https://github.com/akshaykokane/implementing-agentic-rag-with-semantic-kernel-and-azure-ai-search/blob/main/README.md

An example of a POST request to the /rag/agentic API endpoint, showing how to provide a prompt for the Agentic RAG service to plan a trip.

```json
POST /rag/agentic
{
  "prompt": "Plan a weekend trip for family to Acadia National Park"
}
```

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

### Generate Response using Traditional RAG in C#

Source: https://github.com/akshaykokane/implementing-agentic-rag-with-semantic-kernel-and-azure-ai-search/blob/main/README.md

C# code snippet demonstrating how to call the `GenerateResponseUsingTraditionalRag` method on a `ragService` object to get a response based on a user query.

```csharp
var result = await ragService.GenerateResponseUsingTraditionalRag("Best hotels in Bar Harbor?");
Console.WriteLine(result);
```

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

### Generate Response using Agentic RAG in C#

Source: https://github.com/akshaykokane/implementing-agentic-rag-with-semantic-kernel-and-azure-ai-search/blob/main/README.md

C# code snippet illustrating how to use the `GenerateResponseUsingAgenticRag` method with a `ragService` object to get a response that involves agentic reasoning and tool invocation.

```csharp
var result = await ragService.GenerateResponseUsingAgenticRag("What's the weather like for a trip to Acadia National Park?");
Console.WriteLine(result);
```

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

### Run Project using .NET CLI

Source: https://github.com/akshaykokane/implementing-agentic-rag-with-semantic-kernel-and-azure-ai-search/blob/main/README.md

Instructions for cloning the repository, navigating to the project directory, restoring dependencies, building, and running the .NET application using the .NET CLI.

```bash
git clone https://github.com/your-username/AgenticRAG.git
cd AgenticRAG
dotnet restore
dotnet build
dotnet run
```

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

### Configure Azure Services in launchSettings.json

Source: https://github.com/akshaykokane/implementing-agentic-rag-with-semantic-kernel-and-azure-ai-search/blob/main/README.md

This snippet shows the environment variables required in launchSettings.json to configure Azure OpenAI and Azure AI Search services, including resource names, keys, and deployment names.

```json
      "environmentVariables": {
        "ASPNETCORE_ENVIRONMENT": "Development",
        "AZURE_OPENAI_URL": "https://<RESOURCENAME>.openai.azure.com",
        "AZURE_OPENAI_KEY": "<ADD_KEY>",
        "AZURE_AI_SEARCH_URL": "https://<RESOURCENAME>.search.windows.net",
        "AZURE_AI_SEARCH_KEY": "<ADD_KEY>",
        "EMBEDDING_DEPLOYMENT_NAME": "<REPLACE_WITH_DEPLOYMENT_NAME>",
        "GPT_DEPLOYMENT_NAME": "<LLM_MODEL_DEPLOYMENT_NAME>",
        "INDEX_NAME": "<AI_SEARCH_INDEX_NAME>"
      },
```

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

### Import Weather Plugin in Semantic Kernel

Source: https://github.com/akshaykokane/implementing-agentic-rag-with-semantic-kernel-and-azure-ai-search/blob/main/README.md

C# code demonstrating how to import a 'WeatherPlugin' into the Semantic Kernel, making its functionality available for agentic workflows.

```csharp
kernel.ImportPluginFromType<WeatherPlugin>("searchWeather");
```

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

### Send Traditional RAG Prompt

Source: https://github.com/akshaykokane/implementing-agentic-rag-with-semantic-kernel-and-azure-ai-search/blob/main/AgenticRAG/AgenticRAG/wwwroot/index.html

Sends a user prompt to the traditional RAG endpoint. It handles user input, displays messages in a chat interface, makes a POST request to the '/traditionalRag' endpoint, and processes the response or errors. Dependencies include DOM manipulation for input and chat elements.

```JavaScript
async function sendTraditional() {
    const input = document.getElementById('traditional-input');
    const chat = document.getElementById('traditional-chat');
    const prompt = input.value.trim();
    if (!prompt) return;
    chat.innerHTML += `<div><strong>You:</strong> ${prompt}</div>`;
    input.value = "";
    try {
        const response = await fetch('https://localhost:7037/traditionalRag', {
            method: 'POST',
            headers: {
                'Content-Type': 'application/json'
            },
            body: JSON.stringify({ prompt })
        });
        if (!response.ok) {
            const errorData = await response.json();
            chat.innerHTML += `<div class="text-danger"><strong>Error:</strong> ${errorData.message || 'Something went wrong'}</div>`;
        } else {
            const data = await response.json();
            chat.innerHTML += `<div class="text-primary"><strong>RAG:</strong> ${data.content}</div>`;
        }
    } catch (error) {
        console.error("Fetch error:", error);
        chat.innerHTML += `<div class="text-danger"><strong>Error:</strong> Could not connect to the server.</div>`;
    } finally {
        chat.scrollTop = chat.scrollHeight;
    }
}
```

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

### Send Agentic RAG Prompt

Source: https://github.com/akshaykokane/implementing-agentic-rag-with-semantic-kernel-and-azure-ai-search/blob/main/AgenticRAG/AgenticRAG/wwwroot/index.html

Sends a user prompt to the agentic RAG endpoint. This function is similar to sendTraditional but targets the '/agenticRag' endpoint and displays responses with a different styling. It manages user input, chat display, network requests, and error handling.

```JavaScript
async function sendAgentic() {
    const input = document.getElementById('agentic-input');
    const chat = document.getElementById('agentic-chat');
    const prompt = input.value.trim();
    if (!prompt) return;
    chat.innerHTML += `<div><strong>You:</strong> ${prompt}</div>`;
    input.value = "";
    try {
        const response = await fetch('https://localhost:7037/agenticRag', {
            method: 'POST',
            headers: {
                'Content-Type': 'application/json'
            },
            body: JSON.stringify({ prompt })
        });
        if (!response.ok) {
            const errorData = await response.json();
            chat.innerHTML += `<div class="text-danger"><strong>Error:</strong> ${errorData.message || 'Something went wrong'}</div>`;
        } else {
            const data = await response.json();
            chat.innerHTML += `<div class="text-success"><strong>Agentic RAG:</strong> ${data.content}</div>`;
        }
    } catch (error) {
        console.error("Fetch error:", error);
        chat.innerHTML += `<div class="text-danger"><strong>Error:</strong> Could not connect to the server.</div>`;
    } finally {
        chat.scrollTop = chat.scrollHeight;
    }
}
```

=== COMPLETE CONTENT === This response contains all available snippets from this library. No additional content exists. Do not make further requests.