Adding data sources in BMC HelixGPT

To get started with BMC HelixGPT for the chatbot, knowledge article search, and summarization use cases, as an administrator, you must first define the data sources in the HelixGPT Agent Studio. You must define the required parameters for the new data sources to connect the data source with the database into which data will be ingested.

Data sources for BMC HelixGPT

The following data sources are available out-of-the-box in BMC HelixGPT. You can only edit the connection details for these data sources. You cannot delete these connections.:

• BMC Helix Business Workflows
• BMC Helix Knowledge Management by ComAround
• BMC Helix ITSM: Knowledge Management

The following data sources are not available out-of-the-box.  You must first enable them to create new data connections:

• BMC Helix Customer Service Management 
• Confluence 
• Microsoft SharePoint Online 
• Web
• Salesforce Knowledge 

You can add multiple connections for Confluence, Microsoft SharePoint Online, Salesforce Knowledge, and Web data sources.

Before you begin

Make sure you have the BMC HelixGPT Administrator role to add data or edit sources and data connections. Learn more about roles in Roles-and-permissions.

Process for setting up BMC HelixGPT

The following image shows the process of setting up BMC HelixGPT and the current step that you are on:

Complete the following steps to configure data sources in BMC HelixGPT:

• Review the existing out-of-the-box connections
• Enable data sources from record definitions
• Add data sources to BMC HelixGPT
• Create multiple connections for data sources

To review and edit an out-of-the-box data connection

Out-of-the-box data connections are available for BMC Helix Business Workflows, BMC Helix Knowledge Management by ComAround,BMC Helix ITSM: Knowledge Management, and BMC Helix ITSM.You cannot delete these connections and can only edit the details. 

1. In HelixGPT Agent Studio, click Settings .
2. Select HelixGPT > Connections > Information sources.
3. To open a connection, click the connection name.
4. In the Edit Connection pane, modify the following details:
   • Connection name
   • User name
5. Click Save.

To enable data sources

1. As a HelixGPT Agent Studio Admin user, log in to BMC Helix Innovation Studio.
2. On the Workspace tab, click HelixGPT Agent Studio. 
3. On the Records tab, select the check box for the DataSource record definition and click Edit data, as shown in the following image:
   
4. On the Data editor (DataSource) page, select the checkbox of the Microsoft SharePoint Online data source, and click Edit as shown in the following image:
   
5. In the Edit record pane, change the status to Enabled and click Save.
6. Repeat steps 4 and 5 to enable Confluence, Helix Customer Service Management, Salesforce Knowledge, and Web data sources.

(Optional) To register an application in Microsoft Entra ID

Click here to view the steps to register an application in Microsoft Entra ID

1. Log in to the Microsoft Entra admin center.
2. Open  Entra ID > App registrations.
3. Select New registration. 
   The following IDs are created:
   1. Client ID
   2. Tenant ID
4. To create a client secret, click Certificates & secrets.
5. Click New client secret and enter a name.
   A client secret is generated.
6. Click API permissions.
   1. Click Add permissions.
   2. Select Microsoft Graph > Delegated permissions.
   3. Select the following options:
      • openid
      • profile
      • Sites.Read.All
      • User.Read
   4. Select the following application permissions:
      • Files.Read.All
      • Sites.Read.All
      • User.Read.All
7. Click the Grant admin consent for BMC option.

(Optional) To register an application for Microsoft SharePoint Online connection

Click here to view the steps to register an application for Microsoft SharePoint Online connection

1. Navigate to https://portal.azure.com.
2. Click the left navigation pane, select App registrations > New registration to register a new application, and name the application.
   
3. For Supported account types, select Accounts in this organizational directory only.
4. Click Register.
   The Overview section provides two critical pieces of information required for the connector setup. Note both values.

   Field	Description
   Application (client) ID	Used as the Client ID in the Connection_SharePoint form
   Directory (tenant) ID	Used as the Tenant ID in the Connection_SharePoint form
   Display name	Used as the Name in the Connection_SharePoint form

5. Select Client credentials > Add a certificate or secret, and click New client Secret .
   
6. Enter a description and expiration date. Note the Value of the secret, which is required for the Client Secret in the Connection_SharePoint form.
7. In the left menu bar, select API permissions, and click Add a permission.
8. Click API permissions and click Microsoft Graph.
   
9. Click Application permissions and select the following application permissions:
   1. Sites.Read.All
   2. Sites.Selected
      ​​​​​
      
      
10. Click Grant admin consent for to grant the application permissions for the organization.
    
11. Navigate to https://myapps.microsoft.com.
12. Select SharePoint .
    This redirects you to a SharePoint site associated with the account.
13. On the left side, select My Sites, and select an existing site, or create a new one. After creating, select it.
    This redirects you to a URL similar to domain.sharepoint.com/sites/SiteName. This is the value stored in the Site value in the connector and doesn't include the https:.
    
14. After you have collected all the information, select IS > HelixGPT Manager > Connection_SharePoint.
15. Enter the following details and click Save:
    1. Client ID
    2. Client Secret
    3. Name
    4. Site
    5. Tenant ID
    6. Tenant Name
       
16. After the record is created, click it and get the Connection ID.
    You can now create a DataConnectionJob with a Datasource SPT​​​​​​, and the Connection ID that you just created. This job crawls the documents or pages on that SharePoint Site.

To add data sources to BMC HelixGPT

After enabling the data sources, you can add a new data connection in HelixGPT Agent Studio.

1. In HelixGPT Agent Studio, click Settings .
2. Select HelixGPT > Connections > Information sources.
3. To add a new connection, click + Add connection.
4. From the Connection type list, select a connection type.
5. Click Add.

6. In the Add connection pane, specify connection details for the data source you plan to use.

   For more information about connection details, see Data sources for BMC HelixGPT.

7. Click Save.

To edit a data connection

1. In HelixGPT Agent Studio , click Settings .
2. Select HelixGPT > Connections > Information sources.
3. Select the connection name that you want to edit.
4. In the Edit Connection pane, modify the following details:
   • Connection name
   • Confluence- URL, Page ID, Access Token
   • SharePoint- Client ID, Client Secret, Site, Tenant ID, Tenant
   • Helix Customer Service Management- User name
   • Web- URL
   • Salesforce Knowledge- URL, Security Token, User name, Password
5. Click Save.

Results

After the connection is successfully saved, a connection ID is generated. To note the Connection ID for the new data sources that you created, in the data connections table, from the Visible columns list, select ID.

The following image shows the Connection ID in the data sources table:

Use the connection ID while creating the data connection jobs to ingest data into the BMC HelixGPT database. You must create a new job for every Confluence, Microsoft SharePoint Online, Web, or Salesforce Knowledge data connection that you add.

Where to go from here

Ingesting-data-into-BMC-HelixGPT

Related topics

BMC HelixGPT architecture

Setting up and going live