Overview

An AI Assistant refers to the virtual agent or chatbot available within the Space to assist users with their queries and provide relevant information.

An Assistant can contain one or more Modules. Each Assistant should be configured for a specific use case, meaning each use case will likely require a unique set of Modules. The different use cases are described in more detail here.

When a user submits a prompt, the Assistant will determine which Module is best suited to generate a response.

What are Modules?

Modules are building blocks of an Assistant, each providing specific functionalities. The functionality of individual modules is illustrated here.

How are Modules created?

Provided Modules: Unique offers a set of pre-built Modules that clients can use.
Custom Modules: Clients can also develop their own Modules using the Unique SDK.

Important Considerations

Module Compatibility: Not all Modules are universally compatible, meaning not all combinations result in good chat quality. Depending on the use case of the AI Assistant, certain Modules must be combined due to their configuration dependencies. The diagram below shows the combination of required and optional Modules for different use cases. Optional Modules can be integrated but are not essential.

Optimized Configuration: As the number of Modules in an Assistant increases, the likelihood of selecting the most suitable module decreases. When the Assistant has fewer modules to choose from, it is more likely to select the correct one for a given user input. Therefore, it is recommended to keep the number of Modules in an Assistant to a minimum.

How to

Configure a Space’s AI Assistant

Add Modules: When creating or editing a Space, configure the Assistant by finding and selecting modules to add. By clicking on Find AI modules to add, a drop-down list appears with the available modules.
Override Default Configurations and Tool Definition: The default configurations of a Module can be overridden. Overridden changes only apply to the Space being configured. The description of the module can also be customised in the Tool Definition field. This description is used by the assistant when selecting the most suitable module for the user request. Note: Changes to the configuration and tool description affect the performance of the module. Only make changes if you are familiar with the functionality of the module.
Specify a Fallback Module: The Fallback AI Module needs to be specified in the scenario the Assistant is not able to select the best-fitting module given a user’s prompt.
Export and Import Functionality: Assistant configurations can be exported and imported. This feature is particularly useful if you want to use the same configurations in another environment.