SharePoint Connector — Unique AI Documentation

SharePoint Connector

Overview

The SharePoint Connector is a cloud-native application that automatically synchronizes documents from SharePoint Online to the Unique knowledge base. This guide provides administrators with essential information about requirements, features, and limitations.

For deployment, configuration, and operational details, see the IT Operator Guide.

Quick Summary

What it does: Automatically syncs flagged documents from SharePoint Online to Unique's AI knowledge base
Deployment: Kubernetes-based containerized application
Authentication: Uses Azure AD application permissions (no user accounts needed)
Scheduling: Configurable automated scans (default: every 15 minutes)

Requirements

Microsoft 365 / SharePoint Online

Requirement Details
SharePoint Online Active tenant with document libraries
Azure AD Tenant with Global Administrator or Application Administrator rights
License Standard SharePoint Online license (included in most Microsoft 365 plans)

Prerequisites:

Permissions

Content sync: Microsoft Graph:

Permission sync (only if permission sync is enabled): Microsoft Graph:

SharePoint REST API:

For detailed permission justifications, see Microsoft Graph Permissions.

Features

Core Capabilities

Automated Periodic Document Synchronization

Sync Modes

Permission Sync

Selective Sync

Ingestion Modes

Multiple Document Formats

Automatic Change Detection

Flexible Authentication

Advanced Features

Security

Dynamic Site Configuration

Observability

Configuration

How It Works

High-Level Flow

See Architecture Documentation for detailed component diagrams.

Content Sync Flow

See Content Sync Flow for additional details.

User Workflow

  1. Administrator Setup (One-time) - Deploy connector to Kubernetes - Register Azure AD application - Grant site-specific permissions - Configure sync column name
  2. SharePoint Administrator (Per Library) - Add sync column to document libraries - Set column visibility for users
  3. End Users (Ongoing) - Check sync column for documents to sync - Connector automatically picks up flagged files
  4. Automated Processing - Connector scans sites every 15 minutes (configurable) - Downloads flagged files - Uploads to Unique knowledge base - Updates metadata

Limitations and Constraints

Not Supported

Considerations

Constraint Impact Mitigation
Files must be explicitly flagged No automatic sync Document user workflow clearly
Requires custom column per library Setup overhead per library Provide setup guide
One connector per SharePoint tenant Multiple deployments for multi-tenant Document multi-origin workaround
Horizontal scaling not supported Single pod deployment Adequate resource allocation
Singleton per site A site must not be configured in more than one connector process Ensure non-overlapping sites across multiple processes for the same Unique instance
Permission sync group visibility App principal needs member/owner or Full Control Document requirement

Future Versions

Planned enhancements:

Enhancement Rationale
Multi-SharePoint / Multi-Microsoft-Tenant Support High priority on roadmap
Admin UI for configuration Needs feature parity first
On-demand syncing mechanics Business case not yet clear

Related Documentation

For IT Operators

Technical Reference

Standard References