Data Transmission Service:Connect SharePoint to a DTS RAGFlow knowledge base

Step 1: Register an application in Microsoft Entra ID

1. Sign in to the Azure portal.

2. In the left navigation pane, select Microsoft Entra ID.

3. Click App registrations > New registration.

4. Enter the application information, and then click Register.

   Parameter	Description
   Name	Enter an application name, such as KBSync-SharePoint-Reader.
   Supported account types	Select Accounts in this organizational directory only (Single tenant).
   Redirect URI	You can leave this blank. The client credentials flow does not require a redirect URI.

5. After registration, record the following information from the application's Overview page.

   Parameter	Description
   Application (client) ID	The unique identifier of the application, corresponding to the llamahubReader_client_id parameter in the KBSync configuration.
   Directory (tenant) ID	The Microsoft Entra tenant ID, which corresponds to the llamahubReader_tenant_id parameter in the KBSync configuration.

Step 2: Create a client secret

1. On the application's page, select Certificates & secrets > Client secrets.

2. Click New client secret.

3. Enter a description and select an expiration period (we recommend 24 months or Never).

4. Click Add, then immediately copy the secret's value.

   Important

   You cannot view this key value again after you leave the page. Be sure to copy and save it securely immediately after it is created. This key corresponds to the llamahubReader_client_secret parameter in the KBSync configuration.

Step 3: Configure API permissions

1. On the application's page, select API permissions > Add a permission.

2. Select Microsoft Graph > Application permissions.

3. Search for and add the following permissions.

   Permission type	Permission name	Description
   Application	Sites.Read.All or Sites.Selected	Sites.Read.All: Reads all SharePoint sites (Recommended for full site access). Sites.Selected: Reads only selected SharePoint sites (a more secure option). Note If you select the Sites.Selected permission, you must explicitly grant access to specific SharePoint sites for your application via the Microsoft Graph API.
   Application	Files.Read.All	Allows the application to read all files in all site collections.
   Application	BrowserSiteLists.Read.All	Allows the application to read browser site lists.

4. Grant admin consent.

   After adding permissions, a global administrator must grant consent to activate them.

   1. On the API permissions page, click Grant admin consent for <tenant name>.

   2. Confirm that the Status column shows Granted for the permissions.

   Note

   If you do not have administrator privileges, contact your organization's Microsoft 365 global administrator to complete this step.

Step 4: Get SharePoint site information

Before configuring KBSync, collect the following SharePoint information.

Step 1: Prepare the configuration file

1. Obtain the KBSync program and prepare its runtime environment.

   Note

   • Submit a ticket to obtain the KBSync file.

   • The KBSync program requires a Linux environment with access to both the Microsoft Graph API and RAGFlow.

2. Prepare the configuration file named config.

   1. Create a file named config in your Linux environment.

   2. Copy the following code into the config file.

      # Basic configuration
      sourceType=LlamaHub
      sinkType=RagFlowV2
      whiteList=
      blackList=
      
      sleepTime=600
      jobId=dts_kbsync_llamahub_sharepoint_to_ragflow
      documentSyncMode=full
      
      # LlamaHub Reader configuration
      llamahubReaderClass=llama_index.readers.microsoft_sharepoint.SharePointReader
      llamahubReader_client_id=96b75717-****-****-****-46b70e29bec1
      llamahubReader_tenant_id=c2211d60-****-****-****-fd062b3f8c2b
      llamahubReader_client_secret=Xwd8Q***************DmpRTDmpRTw
      llamahubReader_sharepoint_host_name=contoso.sharepoint.com
      llamahubReader_sharepoint_site_name=Marketing
      llamahubReader_drive_name=Documents
      llamahubLoad_sharepoint_folder_path=Reports
      
      # Sink RAGFlow configuration
      sinkOSSAccessKeyId=
      sinkOSSAccessKeySecret=
      sinkOSSRegion=
      sinkOSSBucket=
      sinkOSSEndpoint=

   3. Update the parameters in the config file.

      Important

      • If a parameter is not required for your setup, you can leave its value empty.

      • The parameter blackList has a higher priority than the parameter whiteList.

      Parameter	Required	Description	How to obtain
      whiteList	No	Specifies paths to include (whitelist) or exclude (blacklist) from the synchronization. Note Supports regular expressions. Separate multiple paths with spaces.	Obtain the target folder paths from SharePoint.
      blackList	No
      sourceType	Yes	The type of the source.	Set the value to LlamaHub.
      sinkType	Yes	The type of the sink.	Set the value to RagFlowV2.
      sleepTime	Yes	The interval in seconds for incremental scans.	-
      documentSyncMode	Yes	The sync mode: full: Full sync inc: Incremental sync	-
      llamahubReader_client_id	Yes	The Application (client) ID of the app registered in Microsoft Entra ID.	See Step 1: Register an application in Microsoft Entra ID.
      llamahubReader_tenant_id	Yes	The Directory (tenant) ID from Microsoft Entra ID.
      llamahubReader_client_secret	Yes	The client secret created in Microsoft Entra ID.	See Step 2: Create a client secret.
      llamahubReader_sharepoint_site_name	Yes	The name of the SharePoint site.	See Step 4: Get SharePoint site information.
      llamahubReader_sharepoint_host_name	No Note Required when you use the Sites.Selected permission.	The SharePoint host name.
      llamahubReader_drive_name	No	The name of the SharePoint document library. The default value is Documents.
      llamahubLoad_sharepoint_folder_path	No	The path to a folder within the document library. If left blank, the system syncs all content in the root of the document library.
      sinkOSSAccessKeyId	Yes	Information related to the Object Storage Service (OSS) bucket.	Obtain this information from the Object Storage Service (OSS) console.
      sinkOSSAccessKeySecret	Yes
      sinkOSSRegion	Yes
      sinkOSSBucket	Yes
      sinkOSSEndpoint	Yes

Step 2: Run the KBSync program

1. Place the KBSync file and the config file in the same directory in your Linux environment.

2. In a Linux environment, run the ./KBSync --config config command to start the KBSync program.