Sharepoint Mcp

Name: Sharepoint Mcp
Author: ravikant1918

@ravikant1918

关于 Sharepoint Mcp

Production-grade MCP Server for Microsoft SharePoint — give your AI agent (Claude, Cursor, Continue) live access to read, write, and manage SharePoint files and folders.

#mcp #model-context-protocol #sharepoint #microsoft-graph #ai-agents #enterprise #claude #cursor #vs-code #automation #knowledge-base #document-management #llm-integration #python #docker

基本信息

分类

其他

许可证

MIT

运行时

python

传输方式

stdio

发布者

ravikant1918

提交者

Ravi Kant Yadav

配置

使用下面的配置,将此服务器添加到你的 MCP 客户端。

{
  "mcpServers": {
    "sharepoint": {
      "command": "sharepoint-mcp",
      "env": {
        "SHP_ID_APP": "your-app-id",
        "SHP_ID_APP_SECRET": "your-app-secret",
        "SHP_SITE_URL": "https://your-tenant.sharepoint.com/sites/your-site",
        "SHP_TENANT_ID": "your-tenant-id",
        "SHP_DOC_LIBRARY": "Shared Documents/your-folder"
      }
    }
  }
}

工具

未检测到工具

工具是从 README 中自动提取的。维护者可以在 ## Tools 标题下列出工具,即可填充这部分内容。

概览

What is Sharepoint Mcp?

Sharepoint Mcp is a production-grade Model Context Protocol (MCP) server that gives AI agents live access to Microsoft SharePoint. It connects tools like Claude Desktop, VS Code Copilot, Cursor, and Continue to SharePoint documents, folders, and metadata, enabling agents to read, manage, and reason over organisational knowledge in real time.

How to use Sharepoint Mcp?

Install via pip install sharepoint-mcp, copy .env.example to .env, fill in your Azure AD credentials (app ID, secret, tenant ID, site URL), and optionally set a subfolder scope with SHP_DOC_LIBRARY. Run sharepoint-mcp for stdio transport, or use Docker with docker compose up -d for HTTP transport. Add it to your MCP client’s configuration file (e.g., claude_desktop_config.json) with the required environment variables.

Key features of Sharepoint Mcp

Dual API support: Office365 REST API or Microsoft Graph API
Full folder management: list, create, delete, recursive tree
Document management: upload, download, update, delete, search
Smart parsing of PDF, Word, Excel, and text files
Native KQL search for semantic file finding
Auto-retry with exponential backoff on throttling

Use cases of Sharepoint Mcp

Summarise a quarterly report from a SharePoint folder using an AI agent
Translate a proposal document and save the new version back to SharePoint
Navigate the document library to find files in a specific subfolder
Extract structured text from Excel sheets for data analysis
Search for documents by keyword across the entire site library

FAQ from Sharepoint Mcp

How do I install and configure Sharepoint Mcp?

Install with pip install sharepoint-mcp, then create a .env file from .env.example and set your SharePoint credentials. Optionally restrict operations to a subfolder with SHP_DOC_LIBRARY. Run sharepoint-mcp for stdio transport or use Docker for HTTP.

What credentials are required to connect to SharePoint?

You need an Azure AD app with client ID, client secret, tenant ID, and your SharePoint site URL. A step-by-step guide is available in docs/azure-setup.md. You can choose between Office365 REST API (default) or Microsoft Graph API by setting SHP_API_TYPE.

What transport modes does Sharepoint Mcp support?

It supports two transport modes: stdio (default, for desktop clients like Claude Desktop and Cursor) and http (for Docker, remote agents, and VS Code Copilot Agent mode). Set the TRANSPORT environment variable to switch.

What file formats can the AI agent read from SharePoint?

The server extracts full text from PDF, Word (.docx/.doc), Excel (.xlsx/.xls), and plain text formats (TXT, JSON, Markdown, HTML, YAML, Python). Images, ZIP, and other binaries are returned as file type plus Base64.

Can I deploy Sharepoint Mcp using Docker?

Yes. Clone the repository, configure .env, and run docker compose up -d. This pulls the latest image from DockerHub (or a specific version) and starts the server with HTTP transport on port 8000. A health check endpoint is available at /health.