> For the complete documentation index, see [llms.txt](https://docs.umbraco.com/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://docs.umbraco.com/ai-in-umbraco/17.latest/extending/providers/creating-a-provider.md).

# Creating a Provider

This guide walks through creating a complete custom AI provider from scratch.

## Prerequisites

* A .NET class library project
* Reference to `Umbraco.AI.Core`
* Understanding of the AI service's API

## Step 1: Create the Project

Create a new class library targeting .NET 10:

{% code title="Terminal" %}

```bash
dotnet new classlib -n MyCompany.Umbraco.AI.MyProvider -f net10.0
cd MyCompany.Umbraco.AI.MyProvider
dotnet add package Umbraco.AI.Core
```

{% endcode %}

## Step 2: Create the Settings Class

Define the configuration your provider needs:

{% code title="MyProviderSettings.cs" %}

```csharp
using System.ComponentModel.DataAnnotations;
using Umbraco.AI.Core.EditableModels;

namespace MyCompany.Umbraco.AI.MyProvider;

public class MyProviderSettings
{
    [AIField(
        Label = "API Key",
        Description = "Your MyProvider API key. Use $Umbraco:AI:Secrets:MyProviderApiKey for a config reference.",
        IsSensitive = true,
        SortOrder = 1)]
    [Required]
    public string? ApiKey { get; set; }

    [AIField(
        Label = "Base URL",
        Description = "API endpoint (leave empty for default)",
        SortOrder = 2)]
    public string? BaseUrl { get; set; }

    [AIField(
        Label = "Organization ID",
        Description = "Optional organization identifier",
        SortOrder = 3)]
    public string? OrganizationId { get; set; }
}
```

{% endcode %}

## Step 3: Create the Chat Capability

Implement the chat capability by extending `AIChatCapabilityBase<TSettings>`:

{% code title="MyProviderChatCapability.cs" %}

```csharp
using Microsoft.Extensions.AI;
using Umbraco.AI.Core.Models;
using Umbraco.AI.Core.Providers;

namespace MyCompany.Umbraco.AI.MyProvider;

public class MyProviderChatCapability : AIChatCapabilityBase<MyProviderSettings>
{
    public MyProviderChatCapability(IAIProvider provider) : base(provider) { }

    protected override IChatClient CreateClient(MyProviderSettings settings, string? modelId)
    {
        // Option 1: Use an existing M.E.AI client
        // return new OpenAIChatClient(settings.ApiKey, modelId ?? "default-model");

        // Option 2: Create your own IChatClient implementation
        return new MyProviderChatClient(
            apiKey: settings.ApiKey,
            baseUrl: settings.BaseUrl,
            modelId: modelId ?? "default-model",
            organizationId: settings.OrganizationId);
    }

    protected override async Task<IReadOnlyList<AIModelDescriptor>> GetModelsAsync(
        MyProviderSettings settings,
        CancellationToken cancellationToken = default)
    {
        // Fetch models from your API, or return a static list
        var models = new List<AIModelDescriptor>
        {
            new(new AIModelRef(Provider.Id, "model-standard"), "Standard Model"),
            new(new AIModelRef(Provider.Id, "model-advanced"), "Advanced Model"),
            new(new AIModelRef(Provider.Id, "model-fast"), "Fast Model")
        };

        return models;
    }
}
```

{% endcode %}

## Step 4: Create the Provider Class

The provider class ties everything together:

{% code title="MyProvider.cs" %}

```csharp
using Umbraco.AI.Core.Providers;

namespace MyCompany.Umbraco.AI.MyProvider;

[AIProvider("myprovider", "My AI Provider")]
public class MyProvider : AIProviderBase<MyProviderSettings>
{
    public MyProvider(IAIProviderInfrastructure infrastructure)
        : base(infrastructure)
    {
        // Register capabilities
        WithCapability<MyProviderChatCapability>();

        // Add more capabilities if supported
        // WithCapability<MyProviderEmbeddingCapability>();
    }
}
```

{% endcode %}

## Step 5: Implement IChatClient (If Needed)

If your AI service doesn't have an existing Microsoft.Extensions.AI (M.E.AI) client, implement `IChatClient`:

{% code title="MyProviderChatClient.cs" %}

```csharp
using Microsoft.Extensions.AI;

namespace MyCompany.Umbraco.AI.MyProvider;

public class MyProviderChatClient : IChatClient
{
    private readonly string _apiKey;
    private readonly string _modelId;
    private readonly HttpClient _httpClient;

    public MyProviderChatClient(
        string apiKey,
        string? baseUrl,
        string modelId,
        string? organizationId)
    {
        _apiKey = apiKey;
        _modelId = modelId;
        _httpClient = new HttpClient
        {
            BaseAddress = new Uri(baseUrl ?? "https://api.myprovider.com")
        };
        _httpClient.DefaultRequestHeaders.Add("Authorization", $"Bearer {apiKey}");

        if (!string.IsNullOrEmpty(organizationId))
        {
            _httpClient.DefaultRequestHeaders.Add("X-Organization-Id", organizationId);
        }
    }

    public ChatClientMetadata Metadata => new("MyProvider", new Uri("https://myprovider.com"), _modelId);

    public async Task<ChatResponse> GetResponseAsync(
        IEnumerable<ChatMessage> messages,
        ChatOptions? options = null,
        CancellationToken cancellationToken = default)
    {
        // Implement API call to your provider
        // Convert messages to your API format
        // Send request and parse response
        // Return ChatResponse

        var requestBody = new
        {
            model = _modelId,
            messages = messages.Select(m => new
            {
                role = m.Role.Value,
                content = m.Text
            })
        };

        var response = await _httpClient.PostAsJsonAsync(
            "/v1/chat/completions",
            requestBody,
            cancellationToken);

        response.EnsureSuccessStatusCode();

        var result = await response.Content.ReadFromJsonAsync<MyProviderResponse>(
            cancellationToken: cancellationToken);

        return new ChatResponse(new ChatMessage(ChatRole.Assistant, result!.Content))
        {
            FinishReason = ChatFinishReason.Stop,
            Usage = new UsageDetails
            {
                InputTokenCount = result.Usage?.InputTokens,
                OutputTokenCount = result.Usage?.OutputTokens
            }
        };
    }

    public IAsyncEnumerable<ChatResponseUpdate> GetStreamingResponseAsync(
        IEnumerable<ChatMessage> messages,
        ChatOptions? options = null,
        CancellationToken cancellationToken = default)
    {
        // Implement streaming if supported
        throw new NotSupportedException("Streaming not implemented");
    }

    public object? GetService(Type serviceType, object? serviceKey = null) => null;

    public void Dispose() => _httpClient.Dispose();
}

// Response model for your API
internal class MyProviderResponse
{
    public string? Content { get; set; }
    public UsageInfo? Usage { get; set; }
}

internal class UsageInfo
{
    public int InputTokens { get; set; }
    public int OutputTokens { get; set; }
}
```

{% endcode %}

## Step 6: Package and Install

Build your provider as a NuGet package:

{% code title="MyCompany.Umbraco.AI.MyProvider.csproj" %}

```xml
<Project Sdk="Microsoft.NET.Sdk">
  <PropertyGroup>
    <TargetFramework>net10.0</TargetFramework>
    <PackageId>MyCompany.Umbraco.AI.MyProvider</PackageId>
    <Version>1.0.0</Version>
    <Description>MyProvider integration for Umbraco.AI</Description>
  </PropertyGroup>

  <ItemGroup>
    <PackageReference Include="Umbraco.AI.Core" Version="*" />
  </ItemGroup>
</Project>
```

{% endcode %}

Install in your Umbraco project:

```bash
dotnet add package MyCompany.Umbraco.AI.MyProvider
```

The provider will be discovered automatically on startup.

## Verification

After installation:

1. Navigate to the **AI** section > **Connections**
2. Click **Create Connection**
3. Your provider should appear in the Select AI Provider dialog
4. Configure settings and save
5. Create a profile using the connection
6. Test with a chat request

## Next Steps

* [Provider Settings](/ai-in-umbraco/17.latest/extending/providers/provider-settings.md) - Learn about `[AIField]` options
* [Chat Capability](/ai-in-umbraco/17.latest/extending/providers/chat-capability.md) - Detailed chat implementation guide
* [Embedding Capability](/ai-in-umbraco/17.latest/extending/providers/embedding-capability.md) - Add embedding support


---

# Agent Instructions
This documentation is published with GitBook. GitBook is the documentation platform designed so that both humans and AI agents can read, navigate, and reason over technical content effectively. Learn more at gitbook.com.

## Querying This Documentation
If you need additional information that is not directly available in this page, you can query the documentation dynamically by asking a question.

Perform an HTTP GET request on the current page URL with the `ask` query parameter, and the optional `goal` query parameter:

```
GET https://docs.umbraco.com/ai-in-umbraco/17.latest/extending/providers/creating-a-provider.md?ask=<question>&goal=<endgoal>
```

`ask` is the immediate question: it should be specific, self-contained, and written in natural language.
`goal` is optional and describes the broader end goal you are ultimately trying to accomplish on behalf of the user. GitBook uses it to tailor the answer towards what is most useful for that goal.

The response will contain a direct answer to the question and relevant excerpts and sources from the documentation.

Use this mechanism when the answer is not explicitly present in the current page, you need clarification or additional context, or you want to retrieve related documentation sections.