Skip to content

yuqiuli2/assistant-toolkit

 
 

Repository files navigation

IBM watsonx Assistant Toolkit

The IBM watsonx Assistant Toolkit aims to help developers & data scientists using IBM watsonx Assistant build, deploy and maintain more effective AI solutions for their businesses.

If this is the first time you are accessing IBM watsonx Assistant features, please access the official product documentation

AI Lifecycle

We believe that if we can help you leverage all of the capabilities available within IBM watsonx Assistant, you will be able to build highly effective AI assistants for your business. To that end, experts from the IBM watsonx Assistant team have created the AI lifecycle guide in addition to the open source Python notebooks Dialog Skill Analysis Notebook, Measure & Analyze Effectiveness Notebooks that are intended to help you build, manage and incrementally improve your assistant's AI capabilities. The team incorporates expertise gained from real world engagements with clients from around the world. We hope it helps you to discover ways to get the most out of your own AI implementation.

Integrations

There are many different ways to integrate with IBM watsonx Assistant. This includes many built in channel integrations such as Webchat and Phone but also extends out to many different types of webhooks that can enable all kinds of use cases. The integrations section in this repo contains many useful assets that can be downloaded and used to enhance the capabilities of your IBM watsonx Assistant instance.

Conversational Skills

Conversational skills are a method to interact with automations in watsonx Orchestrate using assistants created with the Conversation builder. The conversational skills section in this repo contains examples and an SDK to help create code-based conversational skills for use in watsonx Orchestrate. For more information view the watsonx Orchestrate documentation.

About

Toolkit for experimentation with watsonx Assistant

Resources

License

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages

  • JavaScript 53.7%
  • HTML 38.9%
  • SCSS 3.1%
  • Jupyter Notebook 2.2%
  • CSS 1.1%
  • Python 0.9%
  • Dockerfile 0.1%