How to Set Up API Workflows in My AI Front Desk

My AI Front Desk now offers a powerful API Workflow feature that allows your AI receptionist to interact with external APIs in real-time during calls. This tutorial will guide you through setting up and customizing your API Workflows.

What are API Workflows?

API Workflows enable your AI receptionist to:

1. Collect specific information from callers
2. Send this information to an external API
3. Receive data from the API
4. Use the received data in the ongoing conversation

This feature essentially gives your AI receptionist the ability to access and use live data during calls, significantly enhancing its capabilities.

Prerequisites

• An active My AI Front Desk account
• Basic understanding of APIs and JSON
• Access to the API(s) you want to integrate

Step 1: Access the API Workflow Settings

1. Log in to your My AI Front Desk dashboard.
2. Navigate to the API Workflow section in your settings.

Step 2: Create a New API Workflow

1. In the API Workflow section, click on "Add Another API Workflow".
2. You'll see a new workflow card appear.

Step 3: Configure the API Workflow

3.1 Set Workflow Name

1. Enter a memorable name for your workflow in the "Workflow Name" field.Example: "Check Customer Balance"

3.2 Choose Request Method

1. Select the appropriate HTTP method (GET, POST, or PUT) for your API request.Note: Choosing POST or PUT will enable the Request Body section.

3.3 Define Information to Collect

This section allows you to specify what information the AI needs to collect from the caller before making the API request. This information gets sent to your API during the request.

1. Click "Add Another Question" to add a new field.
2. For each field, provide:
   • Variable Name: The name to be used in the API request (e.g., "account_number")
   • Type: Choose from Text, Number, or Yes/No
   • Question to Ask: The question the AI will ask to collect this information (e.g., "What's your account number?")

3.4 Set Request Headers (Optional)

If your API requires specific headers:

1. Toggle on the switch for Request Headers.
2. Click "Add Header" to add a new header.
3. Provide the header name and value for each header.Example: Name: "Authorization", Value: "Bearer your_token_here"

‍

3.5 Configure Request Body (for POST/PUT requests)

If you selected POST or PUT as your request method:

1. Toggle on the switch for Request Body.
2. Enter the JSON body for your request. You can use {{variable_name}} to include information collected from the caller.Example: { "account_number": "{{account_number}}", "check_date": "{{current_date}}" }

‍

3.6 Extract Data from API Response

This section allows you to specify how to extract and use data from the API response.

1. Toggle on the switch for Extract Data from API Response.
2. Click "Add Response Mapping" to add a new mapping.
3. For each mapping, provide:
   • Variable Name: A name for the extracted data (e.g., "balance")
   • JSON Path: The path to extract the specific data from the API response (e.g., "$.account.balance")
   • AI Instruction: How the AI should use this data in the conversation (e.g., "The current balance is {{balance}}")

‍

Step 4: Save Your API Workflow

After configuring all sections of your API Workflow, click the "Save" button at the bottom of the form.

Step 5: Test Your API Workflow

1. Use the testing feature in My AI Front Desk to simulate a call.
2. Go through the process of providing the information your workflow is set to collect.
3. Verify that the AI correctly uses the data received from the API in its responses.

Additional Tips

• You can create multiple API Workflows for different purposes.
• Regularly review and update your API Workflows to ensure they meet your current needs.
• Make sure to handle potential API errors gracefully in your AI's responses.
• Keep your API credentials and sensitive data secure.

By following this tutorial, you'll have successfully set up and customized API Workflows in My AI Front Desk, giving your AI receptionist the ability to access and use real-time data during calls, greatly enhancing its capabilities and the service it provides to your callers.

‍

‍