SharePoint Integration
Connect Cumulus to your Microsoft SharePoint document libraries. Once connected, the AI can search, reference, and answer questions about your organisation’s documents directly — no manual uploads required. Results respect each user’s SharePoint permissions at query time.
SharePoint integration requires a Microsoft 365 account. Users signed in with an email and password cannot connect SharePoint.
What it does
With SharePoint connected, Cumulus can:
- Search your documents for relevant information when you ask a question
- Reference specific files without you having to upload them manually
- Show sources — when the AI uses a SharePoint document, it cites the source alongside the response
- Stay in sync — an initial bulk sync runs on connection, then ongoing delta sync via Microsoft Graph change notifications keeps content fresh
Setting up SharePoint
Only Admins and Owners can connect SharePoint libraries.
Go to Settings → SharePoint
Click Settings (gear icon) in the sidebar and select SharePoint.
Grant admin consent
Click Add Connection. You’ll be redirected to Microsoft to approve the Cumulus integration app for your organisation’s Azure AD tenant. This is a one-time step — subsequent connections reuse the same consent.
Link your Microsoft account
After consent, you’ll be prompted to Link Microsoft Account. This signs you in with your personal Microsoft 365 account so Cumulus can browse your SharePoint sites on your behalf.
Each user must link their own Microsoft account. Files only appear in AI answers for a given user if they have access to that file in SharePoint. This enforces SharePoint’s own permission model at query time — Cumulus never surfaces a document the user couldn’t open themselves.
Select a site
The connection wizard opens at the Sites step. Browse or search your available SharePoint sites and click the one you want to connect.
Select a document library
Choose which document library (drive) within that site to sync. Each library becomes a separate connection.
Choose folders to include or exclude (optional)
Browse the folder tree within the library. You can:
- Select specific folders to include — only those folders (and their descendants) are synced
- Exclude folders — mark specific subfolders to skip, useful for archive folders or content outside your team’s scope
- Leave everything unselected to sync the entire library
Inclusions and exclusions are recursive — they apply to all descendants of the selected folder.
Click Connect Library when ready. The connection is created and an initial bulk sync starts automatically.
Member visibility: Non-admin team members can see that SharePoint is connected and its status (e.g. “Active”), but cannot view connection details, folder selections, or sync progress. Only admins and owners have full access to the SharePoint settings page.
You can add multiple connections — one per document library. Repeat the wizard to add more from the same or different SharePoint sites.
Color-tag auto-discovery
Cumulus can automatically discover and index folders based on their SharePoint folder color tag — no need to manually pick folders site-by-site.
How to set it up:
- Go to Settings → SharePoint and choose a color tag for your organisation (red, orange, yellow, green, teal, blue, or purple)
- In SharePoint, apply that color tag to any folder you want indexed — across any connected site
- On the next sync, Cumulus automatically picks up all folders tagged with that color
Auto-discovered folders are highlighted in green and labelled auto in the folder tree inside Cumulus.
This is particularly useful for large organisations with many sites. Instead of configuring each library individually, just tag the right folders in SharePoint and Cumulus keeps up automatically.
To disable color-tag auto-discovery, set the color to None in SharePoint settings.
Sync types
| Sync type | When it runs |
|---|---|
| Initial bulk sync | Runs automatically when a library is first connected — indexes all files in scope |
| Delta sync | Runs ongoing via Microsoft Graph change notifications — picks up new, modified, and deleted files |
| Manual sync | Trigger anytime by clicking the sync icon on a connection card |
Webhook subscriptions for delta sync renew automatically before they expire, so you don’t need to reconnect libraries to keep them live.
Sync progress
Each connection shows real-time sync progress with file-level detail:
| Metric | What it shows |
|---|---|
| Total files | Number of files in the connected library |
| Completed | Successfully indexed files |
| In progress | Files currently being processed |
| Pending | Files waiting to be processed |
| Failed | Files that encountered errors during sync |
| Percent complete | Overall progress bar |
Expand any connection card to see individual file status — each file shows whether it’s indexed, pending, processing, or failed, along with its chunk count and last sync time.
Connection status badges
| Status | Meaning |
|---|---|
| Active | Connected and syncing normally |
| Indexing… | Initial sync in progress — documents are being processed |
| Syncing… | A manual or scheduled sync is running |
| Pending | Waiting for admin consent to complete |
| Error | Sync failed — check the error message on the connection card |
| Offline | Connection has been disconnected |
Using SharePoint in chat
Once connected, just ask naturally — the AI searches your SharePoint content automatically when relevant:
- “Does our employee handbook cover remote work policies?”
- “What does our Q3 sales report say about Canada?”
- “Find the latest version of our service agreement template”
Be specific about what you’re looking for. The AI searches by relevance — the more context you give, the more accurate the result.
Document sources
When the AI references a SharePoint document, it shows the source in the response — including the site name and library. This helps you verify information and navigate back to the original file.
Removing a connection
To remove a SharePoint library:
- Go to Settings → SharePoint
- Find the connection card
- Click the trash icon and confirm
The library’s documents are removed from the AI’s context immediately.
Related
- File Uploads — manually attach files if you don’t use SharePoint
- Google Drive Integration — connect Google Drive alongside or instead of SharePoint
- Admin Setup — organisation-level integration management