Skip to main content
The SharePoint source syncs files from the document libraries of one SharePoint site into the AI’s training index. You choose the site, and optionally one folder. Each connection uses a Microsoft Entra app that your administrator creates.
Synced files are available to the AI as soon as they arrive, and can appear in customer answers. Connect a site or folder that holds only content meant for customers.

​
What you get

  • Document files. PDF (.pdf), Word (.docx), PowerPoint (.pptx) and Markdown (.md) files become searchable text.
  • Excel workbooks. Each .xlsx file becomes one item, with one table per visible sheet.
  • Daily updates. New and edited files are picked up on each sync.
  • Deletions follow SharePoint. A file deleted or moved out of the synced folder leaves the index after the next sync.
  • Links back to the source. Each item keeps the address of its file in SharePoint.

​
What’s not supported today

  • Modern site pages. Pages built in SharePoint (.aspx) are not synced. Only files in document libraries are.
  • Older Office formats. .doc, .ppt and .xls files are skipped. Save them in the current format first.
  • Scanned PDFs without text. Files with no extractable text are skipped.
  • More than one site per connection. Each OpenCX org connects one site.
  • Hidden Excel sheets. They stay out of the index.
  • Very large workbooks. Workbooks over 10 MB are skipped, and text beyond 500,000 characters per workbook is left out.

Connect SharePoint

Create the app and start the sync.

Troubleshooting

Connection errors and missing files.

Website crawler

For content reachable at a public URL.

Connect a knowledge source

All sources.