Vertex AI: Connect action

You can connect to a Google Cloud Platform (GCP) instance using the Connect action. The GCP manages access to your Google Vertex AI instance.

Prerequisites

Note: The examples and graphics provided on this page are for representation purposes only and may not accurately reflect your specific instance. We do not assume responsibility for their maintenance or accuracy.
  • Create a GCP account (GCP Cloud) and create a GCloud project.
  • Log in to your GCloud account at Google Cloud Project and create a project. If you already have access to a project, you can view it here.
  • Create an OAuth 2.0 Client ID. A client ID is used to identify a single application to Google's OAuth Servers. For more details, see Google connect.
You can connect to a GCP instance using the Connect action from Automation 360. To connect to a GCP, you can use one of the following connection methods:
  • User Account: Enter the GCloud account credentials to establish a connection.
  • Service Account: Enter the GCloud service account key to authenticate between servers using the certificate. To create a service account key, see Creating a service account.
  • Control Room OAuth Connection: Select the previously configured OAuth connection in your Automation Anywhere Control Room. For more information on creating OAuth connection, see Create OAuth connection.

Procedure

  1. Open the Google Cloud console.
  2. At the top-left, click Menu > APIs & Services > Credentials.
  3. Click an application under OAuth 2.0 Client IDs that you want to connect. GCP connect
  4. Copy the values from the Client ID, Client secret, and URI fields.
  5. Create a new bot. For more details, see Create your first bot.
  6. In the Automation Anywhere Control Room, navigate to the Actions pane, select Generative AI > Google, drag Connect, and place it in the canvas.
  7. Connect to the GCP using one of these methods: User Account, Service Account, or Control Room OAuth. Connect using User Account

    To connect using the User Account method:

    1. In the User Email Address field, enter the email address that is used to log in to the GCloud account.
    2. In the Client ID field, paste the Client ID copied from the GCloud. This example uses Insecure string (for stronger security, you can use Credential or Variable).
      Note: You can use one of the following options to specify the User Email Address/Client ID/Client secret/Redirect URI:
      • Credential: To implement stronger security, you can use values in the Credential Vault that contains information about the User Email Address/Client ID/Client secret/Redirect URI.
      • Variable: To implement stronger security, you can use a credential variable that contains information about the User Email Address/Client ID/Client secret/Redirect URI.
      • Insecure string: Enables you to enter the User Email Address/Client ID/Client secret/Redirect URI.
    3. In the Client Secret field, paste the Client Secret copied from the GCloud.
    4. In the Redirect URI field, enter the redirect URI from the GCloud account.
    5. Enter Default as the session name.

    Connect using Service Account

    To connect using the Service Account method:

    1. In the Service Account Key field, enter the GCloud service account key. This example uses Insecure string (for stronger security, you can use Credential or Variable).
    2. In the Create Vertex AI session field, the session name is displayed as Default
      Note: If you do not have access to generate and retrieve the Service Account Key, contact your Google Cloud Admin.

    Connect using Control Room OAuth connection

    To connect using the Control Room OAuth, select an available OAuth connection. For information on setting up an OAuth connection, see Create OAuth connection.