Connect SharePoint
Requirements
Section titled “Requirements”- Your workspace needs a Business or Max license
- You are an administrator in the workspace
- You know the Microsoft Tenant ID of your organization (Azure AD)
- You have access to the Azure Portal to grant admin consent
- The SharePoint Site URL that should be indexed
Connect SharePoint
Section titled “Connect SharePoint”Step 1: Grant admin consent
Section titled “Step 1: Grant admin consent”Before you can create the data source, an administrator of your organization must grant the Octopus crawler app access to SharePoint.
- Open the SharePoint connection dialog (step 2 below)
- Enter the Microsoft Tenant ID
- Click the “Grant admin consent” link in the blue notice area
- Sign in as an Azure AD administrator and grant the permission
- You are redirected back automatically
Note: Admin consent only needs to be granted once per customer tenant. After that, any number of SharePoint sites can be connected.
Step 2: Create the data source
Section titled “Step 2: Create the data source”- Go to Settings, Data sources
- Click ”+ Data source”, then “Connect SharePoint”
- Fill in the fields:
| Field | Description |
|---|---|
| Name | Display name of the data source (e.g. “Company Wiki”, “Management”) |
| Microsoft Tenant ID | The tenant ID of your organization. You can find it under: Azure Portal, Entra ID, Overview, Tenant ID |
| SharePoint Site URL | The URL of the SharePoint site (e.g. https://contoso.sharepoint.com/sites/MySite) |
| Document library | Click “Load libraries” to fetch the available libraries and select one |
- Click “Connect SharePoint”
- The initial indexing starts automatically
Find the Microsoft Tenant ID
Section titled “Find the Microsoft Tenant ID”The tenant ID is a unique identifier of your organization in Microsoft Azure:
- Open the Azure Portal
- Navigate to Entra ID (formerly Azure Active Directory)
- On the Overview page you will find the Tenant ID (a UUID in the format
xxxxxxxx-xxxx-xxxx-xxxx-xxxxxxxxxxxx)
Find the SharePoint Site URL
Section titled “Find the SharePoint Site URL”- Open SharePoint in the browser
- Navigate to the desired site (e.g. via the left navigation)
- Copy the URL from the address bar, it typically looks like this:
https://contoso.sharepoint.com/sites/MySitehttps://contoso.sharepoint.com/teams/MyTeam
Important: Use the URL of the site, not of individual documents or the SharePoint start page (
/_layouts/15/sharepoint.aspx).
What gets indexed?
Section titled “What gets indexed?”- All files in the selected document library
- Subfolders are searched recursively
- If no specific library was selected, the default library of the site is indexed (usually “Shared Documents”)
Scoping the indexing to subsets
Section titled “Scoping the indexing to subsets”If you only want to make part of your SharePoint content available to the AI, the clean solution lies in SharePoint itself: create a dedicated document library (or a dedicated SharePoint site) for exactly the content that should be indexed, and connect this library as a data source. That way you get a clear separation between indexed and non-indexed areas, with their own permissions at the SharePoint level.
A direct path or file-type filter (whitelist/blacklist for specific folders or file extensions) is being prepared as an extension of the data source configuration and will be available in an upcoming release.
Workspace data source
Section titled “Workspace data source”SharePoint data sources are always workspace data sources:
- All users in the workspace can search the files in the chat
- The data source appears in the “Workspace data sources” area on the data sources page
- Only administrators can create and delete SharePoint data sources
Synchronization
Section titled “Synchronization”After the initial indexing, changes are detected automatically:
- New files are indexed
- Changed files are re-indexed
- Deleted files are removed from the index
- Renamed or moved files are updated correctly
SharePoint is connected via webhook, changes are typically detected immediately and processed via delta sync. In addition, a full reconciliation runs every 60 minutes.
Search files in the chat
Section titled “Search files in the chat”After indexing, the SharePoint files are available to all users in the chat:
“What does our travel expense policy say?”
“Summarize the results of the most recent quarterly reports.”
“What rules apply to remote work?”
The AI delivers answers with source references. Through the source reference you can open the original file directly in SharePoint.
Security and privacy
Section titled “Security and privacy”- The crawler app uses a separate app registration in Azure AD that has read-only access to SharePoint files
- The credentials (client ID and secret) are stored encrypted
- The app has no access to emails, calendars, or other Microsoft 365 services
- The connection runs over the client credentials flow, no user account is used for access
Common issues
Section titled “Common issues””Libraries could not be loaded”
Section titled “”Libraries could not be loaded””- Admin consent not granted: Click the link in the connection dialog to grant admin consent
- Wrong tenant ID: Verify the tenant ID in the Azure Portal
- Wrong site URL: Make sure the URL is a valid SharePoint site (e.g.
https://contoso.sharepoint.com/sites/MySite)
“Sync failed”
Section titled ““Sync failed””- Permissions revoked: The admin consent may have been revoked. Grant it again
- Site deleted or renamed: Check whether the SharePoint site still exists