SharePoint MCP Server

DISCLAIMER: This project is not affiliated with, endorsed by, or related to Microsoft Corporation. SharePoint and Microsoft Graph API are trademarks of Microsoft Corporation. This is an independent, community-driven project.

SharePoint MCP Server is a Model Context Protocol (MCP) server that connects LLM applications such as Claude to your SharePoint site via the Microsoft Graph API. Use natural language to query documents, manage lists, upload files, and more — directly from your AI assistant.

Features

Prerequisites

• Python 3.10 or higher
• A SharePoint site with Microsoft 365
• An Azure AD application registration with the required Graph API permissions (see docs/auth_guide.md)

Quickstart

1. Clone and install

git clone https://github.com/DEmodoriGatsuO/sharepoint-mcp.git
cd sharepoint-mcp

python -m venv venv
source venv/bin/activate  # Windows: venv\Scripts\activate

pip install -r requirements.txt

2. Configure

cp .env.example .env
# Edit .env with your Azure AD credentials and SharePoint site URL

Required variables in .env:

3. Verify your setup (optional)

python config_checker.py   # Validate configuration
python auth-diagnostic.py  # Test authentication

4. Start the server

# stdio — default, for Claude Desktop / MCP Inspector
python server.py

# HTTP streamable-http — for web services and Copilot agents
python server.py --transport streamable-http --port 8000

# Docker
docker-compose up

Usage

Claude Desktop

Install the server into Claude Desktop:

mcp install server.py --name "SharePoint Assistant"

Or add it manually to claude_desktop_config.json:

{
  "mcpServers": {
    "sharepoint": {
      "command": "python",
      "args": ["/absolute/path/to/sharepoint-mcp/server.py"],
      "env": {
        "TENANT_ID": "...",
        "CLIENT_ID": "...",
        "CLIENT_SECRET": "...",
        "SITE_URL": "..."
      }
    }
  }
}

MCP Inspector (development)

mcp dev server.py

HTTP Server

# streamable-http (recommended for Copilot agents and web clients)
python server.py --transport streamable-http --host 0.0.0.0 --port 8000

# SSE
python server.py --transport sse --host 0.0.0.0 --port 8000

# Via environment variables
MCP_TRANSPORT=streamable-http MCP_PORT=8000 python server.py

Docker

# Build and start (defaults to streamable-http on port 8000)
docker-compose up

# Or run manually
docker build -t sharepoint-mcp .
docker run --env-file .env -p 8000:8000 sharepoint-mcp

Available Tools

The following MCP tools are exposed to the LLM:

For detailed usage examples and example prompts, see docs/usage.md.

Monitoring and Troubleshooting

Logs

The server writes logs to stdout. Set DEBUG=True in .env to enable verbose logging.

Common Issues

Contributing

Contributions are welcome. Please open an issue first to discuss significant changes. See CONTRIBUTING.md for guidelines.

All contributions must pass the quality checks before merge:

black .       # Formatting
ruff check .  # Linting
pytest        # Tests

License

Released under the MIT License. See LICENSE for details.