Dialog Tree Invocation Skill

Updated 

Overview

The Dialogue Tree Invocationskill allows the dynamic workflow node to transfer the conversation to a predefined dialogue tree. This skill is useful for handling complex interactions that require a structured dialogue flow.

Dialog Tree Invocation Skill Use Cases

  • Handle complex queries: Redirect users to a specific dialogue tree for handling complex queries or troubleshooting steps.

  • Manage Fallback Scenarios: Transfer users to a detailed dialogue tree when the AI cannot understand their input.

  • Simplify Multi-Step Interactions: Manage interactions that require multiple steps and prompts, such as booking appointments or processing returns.

Example

Consider the instance, a user needs help with a technical issue. The user provides the input - "I'm having trouble with my internet connection." Now the dynamic workflow node can use the Dialog Tree Invocation Skill skill to redirect the user to a troubleshooting dialogue tree. As an outcome the user follows the steps in the dialogue tree, and the issue is resolved or escalated as needed.

Configure the Dialog Tree Invocation skill

  1. Within your conversational ai application, click on Skills.

  2. Click on Add Skill and select Dialog Tree Invocation.

  3. Enter the skill Name and Description.

    Note: Ensure you provide precise information about what is required of the skill under the Description field as it is passed as context to the gpt.

  4. Under Dialogue tree in Setup Function, select the name of the Dialogue Tree you wish to transfer to.

    Note: The dialog tree being invoked should not include a user replies node.

  5. Under Set Input Variables Mapping, select the Current Bot Output and Next bot Input. These are input variables that will be passed to the dialogue tree. These can be hardcoded or custom field values. Map the dynamic workflow node output variables to the dialogue tree input variables correctly to ensure seamless data transfer.

  6. Configure the output adapter to pass the context back to the dynamic workflow after the dialogue tree interaction ends.