NeuralTrust 1.0.12

NeuralTrust SDK

The NeuralTrust SDK provides a convenient way to interact with the NeuralTrust API. It allows you to create and send traces asynchronously.

Installation

To install the SDK, add the following package references to your project:

<PackageReference Include="System.Net.Http" Version="4.3.4" />
<PackageReference Include="System.Text.Json" Version="8.0.3" />

Usage

Initialization

First, initialize the NeuralTrustClient with your API key:

using NeuralTrust;

var apiKey = "YOUR_API_KEY";
var client = new NeuralTrustClient(apiKey);

You can also set the baseUrl and timeoutInSeconds if needed:

var client = new NeuralTrustClient(apiKey);

Creating and Sending Traces

To create a new trace:

var trace = client.CreateTrace(
    conversationId: "conv_1234",
    sessionId: "sess_123",
    user: new User { UserId = "user_123" },
    metadata: new Metadata
    {
        UserAgent = "Mozilla/5.0 (Macintosh; Intel Mac OS X 10_15_7) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/122.0.0.0 Safari/537.36",
        Timezone = "America/New_York"
    }
);

To end the trace and send it to the API:

// Simulate a user sending a message and the bot generating a response
trace.Generation("Hello, how are you?");
trace.End("I'm good, thank you!");

// Simulate a user asking for their name and the bot retrieving it
trace.Retrieval("What is my name?");
var retrievalResponse = JsonSerializer.Serialize(new
{
    results = new[]
    {
        new { name = "John Doe", similarity = 0.95 },
        new { name = "Jane Smith", similarity = 0.82 },
        new { name = "John Smith", similarity = 0.78 }
    }
});
trace.End(retrievalResponse);

// Simulate a routing decision based on user input
trace.Router("What is my name?");
var routerResponse = JsonSerializer.Serialize(new
{
    route = "name_agent",
    confidence = 0.95m,
    metadata = new
    {
        intent = "name_inquiry",
        entities = new[] { "name" }
    }
});
trace.End(routerResponse);

Metadata Class

The Metadata class allows you to provide additional information about the user's environment. Here are the available options:

• UserAgent (Optional[string]): The user agent string of the browser. If provided, Os and Browser can be inferred and are not necessary.
• Timezone (Optional[string]): The timezone of the user. If provided, Location can be inferred and is not necessary.
• Browser (Optional[string]): The browser used by the user.
• Device (Optional[string]): The device used by the user.
• Os (Optional[string]): The operating system of the user's device.
• Locale (Optional[string]): The locale of the user.
• Location (Optional[string]): The location of the user.

User Class

The User class allows you to provide information about the user. Here are the available options:

• UserId (Optional[string]): The unique identifier of the user.
• UserEmail (Optional[string]): The email address of the user.
• UserPhone (Optional[string]): The phone number of the user.

Full Example

Here is a full example of how to interact with the SDK:

using System;
using System.Collections.Generic;
using System.Threading.Tasks;
using NeuralTrust;
using NeuralTrustApi;

namespace NeuralTrustExample
{
    class Program
    {
        static async Task Main(string[] args)
        {
            var apiKey = "YOUR_API_KEY";
            var client = new NeuralTrustClient(apiKey);

            // Create a new trace
            var trace = client.CreateTrace(
                conversationId: "1234",
                sessionId: "123",
                user: new User { UserId = "123" },
                metadata: new Metadata
                {
                    UserAgent = "Mozilla/5.0 (Macintosh; Intel Mac OS X 10_15_7) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/122.0.0.0 Safari/537.36",
                    Timezone = "America/New_York"
                }
            );

            // Simulate a user sending a message and the bot generating a response
            trace.Generation("Hello, how are you?");
            trace.End("I'm good, thank you!");
            
            // Simulate a user asking for their name and the bot retrieving it
            trace.Retrieval("What is my name?");
            var retrievalResponse = JsonSerializer.Serialize(new
            {
                results = new[]
                {
                    new { name = "John Doe", similarity = 0.95 },
                    new { name = "Jane Smith", similarity = 0.82 },
                    new { name = "John Smith", similarity = 0.78 }
                }
            });
            trace.End(retrievalResponse);

            // Simulate a routing decision based on user input
            trace.Router("What is my name?");
            var routerResponse = JsonSerializer.Serialize(new
            {
                route = "name_agent",
                confidence = 0.95m,
                metadata = new
                {
                    intent = "name_inquiry",
                    entities = new[] { "name" }
                }
            });
            trace.End(routerResponse);
        }
    }
}

Contributing

Pull requests are welcome. For major changes, please open an issue first to discuss what you would like to change.

License

This project is licensed under the MIT License. See the LICENSE file for details.