AI assistant builder API overview

The AI assistant builder uses the AI assistant REST APIs, and the corresponding SDKs. You can use this API to develop applications that interact with resources from the AI assistant builder.

Client applications

To build a virtual assistant or other client application that communicates with an assistant at run time, use the v2 API. You can develop a customer-facing client that can be deployed for production use, an application that brokers communication between an assistant and another service, or a testing application.

By using the v2 runtime API to communicate with your assistant, your application can take advantage of the following features:

  • Automatic state management. The v2 runtime API manages each session with a user, storing and maintaining all of the context data your assistant needs for a complete conversation.

  • Ease of deploying assistants. In addition to supporting custom clients, an assistant can be easily deployed to popular messaging channels such as Slack and Facebook Messenger.

  • Versioning. You can save a snapshot of your content and link your assistant to that specific version. You can then continue to update your development version without affecting the production assistant.

  • Search capabilities. The v2 runtime API can be used to receive responses from the search integration. When a query is submitted that your assistant can't answer, the search integration can find the best answer from the configured data sources.

For more information, see the AI assistant v2 API reference.

Important:IBM watsonx Orchestrate do not support AI assistant v1 API.

For a list of the available API methods, see API methods summary.