Configure and Test Virtual Agent Hub (VAH) Payloads

Bot Builder includes integrated support for configuring and testing Virtual Agent Hub (VAH) payloads and mock data. This enhancement allows administrators to streamline their testing workflows by eliminating the need to switch between Bot Builder and Virtual Agent Hub.

Configure the VAH Integration

  1. In CXone Mpower, click the app selector and select Bot Builder.
  2. Click the bot you want to work with.
  3. Click Preferences icon, which looks like a gear. in the left icon menu.

  4. On the Settings tab, click VAH Integration.

  5. In the VAH Integration dialog box, click the Slot used by VAH dropdown and select the appropriate slot.

  6. Click the speech bubble icon in the top toolbar to open the chat interface.

  7. In the chat interface, click the curly brackets icon at the bottom-right corner to open the VAH Testing Data panel.

  8. In the VAH Testing Data panel, update the JSON with the sample data you want to test.

  9. Use the chat interface to simulate a conversation and validate the flow behavior using the mock data.

A contact center administrator is building a virtual agent flow that collects a customer’s account number and returns account details. Before deploying the flow, the administrator wants to test how the virtual agent interacts with the VAH using mock data to ensure the payload is correctly configured and the response is as expected.

Using the new VAH integration in Bot Builder, the administrator can:

  1. Select the appropriate VAH slot, for example, accountNumber.

  2. Input mock data in JSON format. For example, inject the account number, CZ650800000019200145399.

    { "name": "Alice Novak", "address": "123 Main Street, Prague, CZ 11000","account_type": "Checking","account_number": "CZ650800000019200145399","status": "Active"

  3. Test the flow directly within the chat interface. For example, simulate the following conversation to test the data,

    User (simulated input):"Hi, I want to check my account balance."

    Agent (Bot Response): "Sure, I can help with that. Can you please provide your account number?"

    User (mock data injected): accountNumber: CZ650800000019200145399

    Agent (Bot Response): "Thanks! I’ve found your account. Your current balance is $1,200.00."

This process helps validate the flow logic and payload structure without switching between multiple tools.