The Processing Node delivers one dynamic response and then automatically ends the conversation. It is best suited for scenarios where a single, concise output is enough to address the user’s request without requiring follow-up.

​

Adding a Processing Node in a Dialogue Tree Tool

1. Add a Dialogue Tree Tool.

2. Click the +Create Dialogue Tree button under Setup Function.

   ​

   ​

3. On the Dialogue Tree builder, click the + icon and add a Processing Node.

   ​

   ​

4. On the Processing Node window, enter the Name, System Prompt, and Input Variable (which is the user prompt for LLMs) as per the relevant use case.

5. Enter the Output Variable: This is the variable where the API Output is stored.

6. Select the Tools from the dropdown. These tools are essential to the Processing Node, allowing the AI to carry out tasks such as updating user information, processing text, making API calls, and routing conversations to human agents.

   ​

   ​

   ​

   Note: The output variable is returned as an array. To use it in a bot reply, reference it in the following format: ${output[0]}.

   

7. Select the LLM Configuration to define how Large Language Models (LLMs) should be configured across different use cases.

   ​

   ​

8. Click Save in the bottom right corner.