SharePoint

Provides MCP access to SharePoint folders, documents, and content processing with text and binary support.

python

GitHub Stars

python

Language

6 months ago

First Indexed

2 months ago

Catalog Refreshed

Documentation & install

Readme and setup notes from the catalogue, plus a client-ready config you can copy for your MCP host.

Installation

Add the following to your MCP client configuration file.

Configuration

View docs

{
  "mcpServers": {
    "sofias-ai-mcp-sharepoint": {
      "command": "python",
      "args": [
        "-m",
        "mcp_sharepoint"
      ],
      "env": {
        "SHP_ID_APP": "YOUR_APP_ID",
        "SHP_SITE_URL": "https://your-tenant.sharepoint.com/sites/your-site",
        "SHP_MAX_DEPTH": "15",
        "SHP_TENANT_ID": "YOUR_TENANT_ID",
        "SHP_DOC_LIBRARY": "Shared Documents/your-folder",
        "SHP_LEVEL_DELAY": "0.5",
        "SHP_ID_APP_SECRET": "YOUR_APP_SECRET",
        "SHP_MAX_FOLDERS_PER_LEVEL": "100"
      }
    }
  }
}

You can connect your MCP clients to SharePoint resources through a dedicated MCP Server that handles folders, documents, and content processing. It supports text and binary files, with intelligent extraction for common document formats, making it easy to manage SharePoint content from the MCP ecosystem.

How to use

You expose SharePoint resources to MCP clients via a local or remote MCP server instance. Start the server using the recommended runtime, then connect your MCP client to the server to browse folders, list documents, upload or update files, and retrieve content. The server automatically detects content type and provides text extraction for PDFs, Word, Excel, and other text formats.

How to install

Prerequisites: Python 3.10 or newer is required to run the server locally. You should also have access to a SharePoint site and the appropriate Azure AD app credentials.

Install the MCP SharePoint server package in editable mode during development or from a published package.

Install from source (editable):

pip install -e .

Install from PyPI when published:

pip install mcp-sharepoint-server

Run the server locally using Python. This starts the MCP server in stdio mode and reads environment variables for configuration.

python -m mcp_sharepoint

Additional sections

Configuration notes and runtime details help you tailor the server to your SharePoint environment and security requirements.

Configuration and runtime notes

Environment variables shown here are used to connect to your Azure AD app and specific SharePoint site and document library. Provide secure values in your runtime environment.

Available tools

List_SharePoint_Folders

Lists all folders in a specified directory or root to help you navigate the structure.

Create_Folder

Creates new folders in a specified directory on SharePoint.

Delete_Folder

Safely deletes an empty folder from SharePoint.

Get_SharePoint_Tree

Provides a recursive tree view of the SharePoint folder structure with configurable depth.

List_SharePoint_Documents

Fetches all documents within a specified folder along with metadata.

Get_Document_Content

Retrieves and processes document content, including text extraction from PDFs, Word, and Excel files.

Upload_Document

Uploads new documents to a specified folder and handles both text and binary content.

Upload_Document_From_Path

Uploads a local file directly from the filesystem for large documents.

Update_Document

Updates the content of existing documents in SharePoint.

Delete_Document

Removes documents from specified folders.

Excel_Content_Processing

Extracts data from all sheets in Excel files and converts to readable text (first 50 rows per sheet).

Word_Content_Processing

Processes Word documents to preserve paragraphs and tables with structure.

PDF_Content_Processing

Performs full text extraction from PDFs using PyMuPDF for accurate results.

Text_Content_Processing

Processes text-based formats (JSON, XML, HTML, MD, code) for easy consumption.

Binary_Support

Base64 encoding/decoding for seamless handling of binary files.

Built by

VeilStrat

AI signals for GTM teams