Share point api/plugin connect - Site/List/Document Library

HR_T_EN

I would like to know the step by step process to follow for me to access contents present in one of my sharepoint sites (.aspx) and also to connect to a sharepoint online list and documents present in default library to build a knowledge base for my RAG based agent.

Dataiku version used: 14.7.0

ThomasW

Hello,

To build a knowledge base for a RAG-based agent from SharePoint Online content in DSS, the process would be:

1. Create a SharePoint Online connection in DSS and configure the Microsoft authentication settings. For the detailed setup steps, please refer to the SharePoint Online connector documentation:
   SharePoint Online connector documentation
2. In your flow > click + ADD ITEM button and create either:
   • a SharePoint List dataset, or
   • a SharePoint document dataset / managed folder
3. Build a knowledge bank from that content.
4. Use the knowledge bank in your RAG-based agent. For the detailed knowledge bank and agent setup steps, you can follow this tutorial:
   RAG conversational agent tutorial

Please note that SharePoint site pages are not currently supported. If the content exists only as SharePoint pages, an additional workaround or custom extraction step would likely be needed before building the knowledge base.

Best,