Text user input control
-
- UpdatedNov 7, 2024
- 7 minutes to read
- Xanadu
- Virtual Agent
The Text user input control in a Virtual Agent topic prompts the user for a text string.
Text user input control properties for LLM topic discovery
Property | Description |
---|---|
Node name |
Name that identifies this node in the topic flow. |
Variable name |
Name of the variable that stores the user response to this prompt. The variable name is automatically created from the Node name property. |
Allow automatic slot-filling | Toggle to allow this node’s value to be automatically set by other nodes, as well as during topic discovery. When this toggle is activated, the Detail description field changes from dynamic Data Pill Picker and script options to a static text field. |
Detail description |
Description of the information that you need from users. If the Allow automatic slot filling toggle is deactivated, you can create a dynamic detail description by
selecting the Data Pill Picker icon |
Input format |
Text format that is validated when a user enters certain text items. If the user doesn't enter the expected format, an error message indicates that the format is not valid and asks the user to
reenter the text. Choose the format of the text item to be validated:
|
Advanced | |
Enforce user prompt |
Toggle to open a field for a custom message shown to the user instead of the message generated by the large language model (LLM). You can create custom messages in plain language, input a script, or define conditions with the data pill picker. |
Validation |
Add a script to validate the user input. For example, check if the user entered a valid email address. |
Additional instructions for LLM |
Further instructions to the LLM, such as adding data formats, restrictions, or default values for user responses. You can create additional instructions in plain language, input a script, or define conditions with the data pill picker. |
Hide or skip this node | |
Conditionally use this node if | No-code condition statement or low-code script that specifies a condition for presenting this node in the conversation. The condition must evaluate to true. |
Allow user to skip this node if |
No-code condition statement or low-code script that specifies a condition for letting users skip this node in the conversation. The condition must evaluate to true. You can set this field using either the condition builder or a script. |
Example Text user input control for LLM topic discovery
Input properties | Input prompt |
---|---|
![]() |
Text user input control properties for NLU topic discovery
Property | Description |
---|---|
Node name |
Name that identifies this node in the topic flow. |
Variable name |
Name of the variable that stores the user response to this prompt. The variable name is automatically created from the Node name property. |
Prompt | Prompt or question for the user. The prompt can be either a text string or a script that returns text. This value is used only when the default value is not specified. For example: What's your name? |
Input format |
Text format that is validated when a user enters certain text items. If the user doesn't enter the expected format, an error message indicates that the format is not valid and asks the user to
reenter the text. Choose the format of the text item to be validated:
|
NLU entity | Option to associate an NLU entity with the node. If an NLU entity is associated with the input variable for this node, Virtual Agent can slot-fill the specified value based on the user's utterance. Select an entity from the list of entities associated with the topic intent. When you specify an entity for the node, the Do not ask users to confirm recognized entity toggle switch is displayed. When enabled, users are not prompted to confirm the extracted entity. |
Advanced | |
Make this input secure | |
Enable | Toggle switch that encrypts the input control. When enabled, the user input is masked and appears as a series of dots in the chat window. |
Hash function [Visible when Make this input secure is enabled] |
The method digest algorithm based on the standard WS-security standard. Choose one of the following:
|
Encryption salt [Visible when Make this input secure is enabled] |
Random data that is used as an additional input to a one-way function that hashes data. |
Default value | |
Predefine a value for user input | Predefined value for the user response to the question or prompt. The response defined in the Default value confirmation field asks the user to confirm the default value. If the user responds with no, the value becomes null. The default value can be either a text string or a script that returns text. For example, if you're using dot-walking, the default value might be: Script Variables > Last username. Or if you're using a script, the default value might be: {{vaScripts.lastUsername}}. |
Confirmation messages | |
Input completion confirmation | Bot response shown to the user when the node interaction is complete. The message can be either a text string or a script that returns text. For example, if you're using dot-walking: Thanks, (Input Variables > Username)! Or if you're using a script, the acknowledgement might be: Thanks, {{vaInputs.username}}! |
Default value confirmation | Message that asks the user to verify that the value in the Default value field is correct. This message is used instead of a value in the Prompt field. It can contain either a text string or a script that returns text. For example, if you're using dot-walking: Are you (Input Variables > Username)? Or if you're using a script, the confirmation message might be: Are you {{vaScripts.lastUsername}}?. |
Conversation switching | |
Turn on to let users change the subject | Option to enable NLU prediction for this node. If enabled, users can enter text to answer questions, regardless of the type of input control being used. Virtual Agent uses this utterance to match another existing intent, letting the user switch topics. |
Hide or skip this node | |
Conditionally show this node if | No-code condition statement or low-code script that specifies a condition for presenting this node in the conversation. The condition must evaluate to true. |
Allow user to skip this node if |
No-code condition statement or low-code script that specifies a condition for letting users skip this node in the conversation. The condition must evaluate to true. You can set this field using either the condition builder or a script. |
Skip reprompting |
No-code condition statement or low-code script that specifies a condition for letting users skip reprompting in the conversation. When a preceding node is revisited through a topic loopback or Dialog Act, the Virtual Agent bypasses this node and automatically retain its original value. |
Example Text user input control for NLU topic discovery
Input properties | Input prompt |
---|---|
![]() ![]() |
![]() |
![]() |