Integration Designer

​

Overview

​

Key features

​

Integration Designer guides

​

Integration with external systems

​

Integration in FlowX.AI

1

Define a Data Source

Navigate to the Integration Designer and create a new data source:

• Name: Airtable Credit Data
• Base URL: https://api.airtable.com/v0/

2

Define Endpoints

In the Endpoints section, add the necessary API endpoints for system integration:

1. Get Records Endpoint:
   • Method: GET
   • Path: /${baseId}/${tableId}
   • Path Parameters: Add the values for the baseId and for the tableId so they will be available in the path.
   • Header Parameters: Authorization Bearer token

See the API docs.

2. Create Records Endpoint:
   • Method: POST
   • Path: /${baseId}/${tableId}
   • Path Parameters: Add the values for the baseId and for the tableId so they will be available in the path.
   • Header Parameters:
     • Content-Type: application/json
     • Authorization Bearer token
   • Body: JSON format containing the fields for the new record. Example:

   {
    "typecast": true,
    "records": [
        {
            "fields": {
                "First Name": "${firstName}",
                "Last Name": "${lastName}",
                "Age": ${age},
                "Gender": "${gender}",
                "Email": "${email}",
                "Phone": "${phone}",
                "Address": "${address}",
                "Occupation": "${occupation}",
                "Monthly Income ($)": ${income},
                "Credit Score": ${creditScore},
                "Credit Status": "${creditStatus}"
            }
        }
    ]
}

3

Design the Workflow

1. Open the Workflow Designer and create a new workflow.
   • Provide a name and description.
2. Configure Workflow Nodes:
   • Start Node: Initialize the workflow.

On the start node add the data that you want to extract from the process. This way when you will add the Start Workflow Integration node action it will be populated with this data.

{
"firstName": "${firstName}",
"lastName": "${lastName}",
"age": ${age},
"gender": "${gender}",
"email": "${email}",
"phone": "${phone}",
"address": "${address}",
"occupation": "${occupation}",
"income": ${income},
"creditScore": ${creditScore},
"creditStatus": "${creditStatus}"
}

Make sure these keys are also mapped in the data model of your process with their corresponding attributes.

• REST Node: Set up API calls:
  • GET Endpoint for fetching records from Airtable.
  • POST Endpoint for creating new records.
• Condition Node: Add logic to handle credit scores (e.g., triggering a warning if the credit score is below 300).

Condition example:

input.responseKey.data.records[0].fields["Credit Score"] < 300

• Script Node: Include custom scripts if needed for processing data (not used in this example).
• End Node: Define the end of the workflow with success or failure outcomes.

4

Link the Workflow to a Process

1. Integrate the workflow into a BPMN process:
   • Open the process diagram and include a User Task and a Receive Message Task.

In this example, we’ll use a User Task because we need to capture user data and send it to our workflow.

2. Map Data in the UI Designer:
   • Create the data model
   • Link data attributes from the data model to form fields, ensuring the user input aligns with the expected parameters.

3. Add a Start Integration Workflow node action:

• Make sure all the input will be captured.

5

Monitor Workflow and Capture Output

Receive Workflow Output:

• Use the Receive Message Task to capture workflow outputs like status or returned data.
• Set up a data stream in the node’s Integration Output section to ensure workflow output is mapped to a predefined key.

One key per hop. Data crosses three boundaries on its way back to the process, and each one uses its own name:

1. Inside the workflow, the REST node stores its result under its Response Key - the body is at ${responseKey.data}.
2. The End node maps that value into the workflow’s declared output parameter.
3. The Receive Message Task maps the returned output into a process attribute through its data stream output mapping.

Most “empty value” problems come from using a key from the wrong hop, or from a typo - a wrong key in a ${...} placeholder resolves silently to nothing rather than raising an error. If the result never reaches the process, check the error glossary for the two runtime errors that pinpoint which hop is broken.

6

Start the integration

• Start your process to initiate the workflow integration. It should add a new user with the details captured in the user task.

• Check if it worked by going to your base in Airtable. You can see, our user has been added.

​

Export and import

​

Exporting

1. Navigate to the Data Sources or Workflows list in Integration Designer
2. Open the context menu for the resource you want to export
3. Select Export
4. The resource is downloaded as a ZIP file containing the full configuration

• Data Sources - data source configuration, endpoints, NoSQL operations, MCP tools
• Workflows - workflow nodes and their configurations

​

Importing

1. Navigate to the Data Sources or Workflows list in Integration Designer
2. Open the context menu and select Import from ZIP
3. Select the ZIP file exported from another project or environment
4. If resources with the same identifiers already exist, a Review Resource Identifiers Conflicts modal appears
5. Choose a strategy for each conflicting resource (or use Apply to all):
   • Keep both - imports the resource as a new copy alongside the existing one
   • Replace - overwrites the existing resource with the imported version
   • Skip this one - keeps the existing resource unchanged
6. Click Continue to complete the import

​

FAQs

{
  "query": "mutation { updatePerson(input: { name: \"${input.name}\" }) { id status } }"
}

​

Academy courses