š¤ About This Assistant
OpenAI GPT-4
Azure OCR
This intelligent assistant uses OpenAI's GPT-4 model with function calling to search and analyze your SharePoint documents in real-time.
š How It Works
- Ask questions naturally - The AI understands context and can search SharePoint automatically
- Document extraction - Reads PDFs, Word docs, Excel files, PowerPoint, and text files
- OCR support - Uses Azure Form Recognizer to extract text from scanned/image-based PDFs
- Conversation memory - Maintains context throughout your session
š” Example Queries
- "What files are in the teaching folder?"
- "Search for prayer updates from September"
- "Summarize the October goal update"
- "Find all documents about website design"
- "What's the latest prayer update?"
āļø Technology Stack
- AI Model: OpenAI GPT-4o with function calling
- Document Processing: pdf-parse, pdf2json, mammoth, exceljs
- OCR: Azure Form Recognizer (for scanned PDFs)
- Data Source: Microsoft SharePoint via Graph API
- Backend: Node.js + Express
š Supported File Types
- š PDF (text-based and scanned/image-based)
- š Word documents (.docx)
- š Excel spreadsheets (.xlsx)
- š½ļø PowerPoint presentations (.pptx)
- š Plain text files (.txt)
š Privacy & Security
- All queries use secure HTTPS connections
- SharePoint access via Azure AD authentication
- Conversation data is session-based and not permanently stored
- Document content is processed on-demand, not cached
Powered by OpenAI GPT-4o ⢠Azure Form Recognizer ⢠Microsoft Graph API
Assistant ID: asst_iTfBChPJuMtDdaRDItRkA39W