DATA-SOURCES.md

Data Sources

Navigation

• 00.OVERVIEW
• Action Planner
• Actions
• AI System
• Application class
• Augmentations
• Data Sources
• Function Calls
• Moderator
• Planner
• Powered by AI
• Prompts
• Streaming
• Turns
• User Authentication

---

Data sources allow the injection of relevant information from external sources into prompts, such as vector databases or cognitive search. A vector data source makes it easy to add RAG to any prompt, allowing for better and more accurate replies from the bot.

Within each Action Planner’s prompt management system, a list of data sources can be registered. For each data source, a max number of tokens to use is specified, via maxTokens.

Customize a Data Source

1. Construct a class that implements our DataSource base class.

2. Register the data source with your prompt management system through the addDataSource function.

3. To augment a specific prompt, you can specify the name(s) of the data sources within the prompt's config.json file.

Our simplest example (primarily for testing) is TextDataSource, which adds a static block of text to a prompt.

Our most complex Javascript example is the VectraDataSource in the Chef Bot sample, which uses an external library called Vectra.

Our most complex C# example is the KernelMemoryDataSource in the Chef Bot sample, which uses an external library called Kernel Memory.

Customized Examples

Javascript

Here is an example of the configuration for the Chef Bot sample:

// Inside VectraDataSource.ts
export class VectraDataSource implements DataSource

// Inside of index.ts
planner.prompts.addDataSource(
    new VectraDataSource({
        name: 'teams-ai',
        apiKey: process.env.OPENAI_KEY!,
        azureApiKey: process.env.AZURE_OPENAI_KEY!,
        azureEndpoint: process.env.AZURE_OPENAI_ENDPOINT!,
        indexFolder: path.join(__dirname, '../index')
    })
);

Inside the prompt's config.json. Here, teams-ai denotes the name of the VectraDataSource, and 1200 is maxTokens.

"augmentation": {
    "augmentation_type": "none",
    "data_sources": {
        "teams-ai": 1200
    }
}

C#

Here is an example of the configuration for the Chef Bot sample:

// Inside KernelMemoryDataSource.cs
public class KernelMemoryDataSource : IDataSource

// Inside of Program.cs
KernelMemoryDataSource dataSource = new("teams-ai", sp.GetService<IKernelMemory>()!);
prompts.AddDataSource("teams-ai", dataSource);

Inside the prompt's config.json. Here, teams-ai denotes the name of the KernelMemoryDataSource, and 900 is the maxTokens.

"augmentation": {
    "augmentation_type": "none",
    "data_sources": {
        "teams-ai": 900
    }
}

The Kernel Memory library provides tools for indexing and querying data. The Chef Bot uses Kernel Memory as an example of how to integrate Retrieval Augmentation (RAG) into the AI library.

---

Return to other major section topics:

• CONCEPTS
• MIGRATION
• QUICKSTART
• SAMPLES
• OTHER