SharepointO365

Use SharepointO365 to synchronize and index content from SharePoint into Content Intelligence, which makes enterprise content searchable and available to AI agents and workflows. Agents and workflows can then use Content Intelligence retrieval functions to search the indexed content and generate answers grounded in it.

Content freshness depends on the latest successful synchronization and indexing cycle.

Choose the Right Connector

Requirement Connector
Synchronize and index SharePoint content into Content Intelligence (CI) so agents and workflows can retrieve relevant documents and passages and generate grounded answers SharepointO365
Use Microsoft Graph APIs to perform operations on SharePoint sites, libraries, lists, files, and pages Microsoft SharePoint
Perform operations within a configured SharePoint site, with support for file browsing and linking Microsoft SharePoint Site

Before You Begin

Have these details ready to configure the SharepointO365 connector:

  • SharePoint URL, site name, and folders to synchronize.
  • Microsoft tenant ID and application client ID.
  • Client private key and matching certificate.
  • Application access to read the selected content and required permission information.
  • Proxy details, if required.

SharePoint Application Permissions

SharepointO365 uses the configured Entra application to read files and metadata from SharePoint during synchronization. In 26D and later, it also reads permission and group-membership information for security synchronization.

Grant the application the required SharePoint API permissions and obtain administrator consent. These permissions support background synchronization; agents access the indexed content through Content Intelligence retrieval functions.

1. Configure the Connection

In AI Agent Studio, go to Connectors → Add Connector. The right panel lists the available out-of-the-box connectors. Select SharepointO365, then enter the instance name, description, family, and product.

Configuration What to provide
Authentication Select Use Certificate and enter both Client Key and Client Certificate.
SharePoint location Enter the SharePoint URL and site name.
Application identity Enter the tenant ID and application client ID.
Folder scope Controls which folders are synchronized and indexed. Enable Start From Root Folder to start from the configured site’s root, or disable it and specify Include Folders to select folders. Use Exclude Folders to omit folders. Folder scope does not grant access or replace user permissions.
Proxy Enable and configure the proxy only when required.

Example source values:

SharePoint URL: https://<tenant>.sharepoint.com
Site Name: <site-name>
Folder Path: /sites/<site-name>/Shared Documents/<folder-name>

Select only the source locations needed for your use case.

2. Configure Access

Release How access is controlled
26C Access is controlled by the Knowledge Management user groups selected for the connector. SharePoint file permissions are not synchronized.
26D New connectors use synchronized SharePoint permissions to control access. Users must have access to the source file, and their identities must align between Fusion and SharePoint.
Note: The application's SharePoint permissions allow the connector to synchronize content. They do not automatically grant users access to that content through agents or workflows.

3. Synchronize and Index

Synchronization runs automatically in the background every 30 minutes. The connector checks SharePoint for new, updated, and deleted files. Changes become searchable after synchronization and indexing is complete.

4. Use the Connector in an Agent or Workflow

  1. In AI Agent Studio, go to Resources → Tools and click Add.
  2. Set Tool Type to Connector.
  3. Select the SharepointO365 connector instance you configured.
  4. Select the connector functions your agent or workflow needs.
  5. Save the tool and add it to the agent in your workflow.

The following connector functions are available:

Function Purpose Available from
get-content-intelligence-search-results Search indexed knowledge 26C
get-content-intelligence-search-chunks Retrieve relevant passages 26C
get-content-intelligence-search-documents Find indexed documents 26C
get-content-intelligence-search-in-document-by-documentId Search within a document 26D
get-content-intelligence-search-in-document-by-versionId Search within a document version 26D
get-content-intelligence-browse-documents Browse indexed documents 26D

Instruct the agent to answer using retrieved content, include source references when available, and explain when sufficient information cannot be found.

For example, an agent can retrieve an order’s status from an indexed spreadsheet and use that information to answer the user’s question.

Troubleshooting

Problem Possible cause
Connection or synchronization fails Incorrect credentials, insufficient application permissions, invalid site details, or network/proxy issues
Expected wrapper articles are missing from Knowledge Authoring Synchronization has not completed, files are outside the configured folder scope, or ingestion failed
Wrapper articles appear in Knowledge Authoring, or the filename can be found, but the file’s contents are not searchable File-content indexing is still processing or has failed
No search results The content is not yet indexed, the user lacks access, or Fusion and SharePoint identities do not match as required
Answers are outdated Recent source changes have not yet completed synchronization and indexing
Sharepoint connector configuration details have changed Delete the Connector with incorrect details and create a new Sharepoint connector instance rather than updating an existing one. 
Files are not getting synchronized or searchable in CI
  1. Check supported file types in config and update as required. Refer Knowledge Management file-type setup.
  2. File size exceeds the limit of 25 MB
  3. Check if file is password protected. Remove password protection and try again.