MCP Server Microsoft365 Filesearch

🚀 MCP Server - Microsoft 365 File Search (SharePoint and OneDrive)

This is an implementation of a Model Context Protocol (MCP) server that offers advanced file search capabilities within the Microsoft 365 environment. It enables efficient file discovery, metadata analysis, and integration with business processes by providing content from SharePoint/OneDrive.

Click here for more details

🚀 Quick Start

✨ Features

• Advanced Search: Conduct high - level file searches in the Microsoft 365 environment.
• Metadata Analysis: Analyze file metadata effectively.
• Business Process Integration: Integrate seamlessly with business processes.
• Local Cache: Improve performance by caching downloaded files locally.

📦 Installation

To use this server, you need to register an application in the Azure Portal:

1. Register a new application.
2. Record the Client ID and Tenant ID.
3. Create a Client Secret under Certificates & secrets.
4. Under API permissions, add the following Delegated or Application permissions:
   • Sites.Read.All
   • Files.Read.All
5. Click Grant admin consent for these permissions.

💻 Usage Examples

Basic Usage

To integrate the server with Claude Desktop, update your claude_desktop_config.json:

"mcpServers": {
  "M365 File Search (SharePoint/OneDrive)": {
    "command": "uv",
    "args": [
      "--directory",
      "full_path_to_parent_directory",
      "run",
      ".\\src\\mcp_m365_filesearch\\server.py"
    ],
    "env": {
      "CLIENT_ID": "MSGraph Client ID",
      "CLIENT_SECRET": "MS Graph Client Secret",
      "TENANT_ID": "Tenant ID",
      "REGION": "Search Region"
    }
  }
}

Advanced Usage

The server provides two core tools:

search_m365_files

# Execute a file search in the Microsoft 365 environment.
# Input:
# - query (str): Search term or condition.
# Return value: An array of file metadata objects. Metadata includes file content summary, *Drive ID* and *File ID*, and other details.

get_file_content

# Retrieve the content of a specific file.
# Input:
# - driveid (str): The unique identifier of the parent drive.
# - fileid (str): The unique identifier of the file.
# Return value: The file content as a binary stream.
# Note: Use local cache to speed up repeated access.

🔧 Technical Details

Components

The server consists of two main components: tools and cache.

Tools

• search_m365_files: This tool is used to execute file searches in the Microsoft 365 environment. It takes a search query as input and returns an array of file metadata objects.
• get_file_content: This tool retrieves the content of a specific file. It uses the drive ID and file ID as input and returns the file content as a binary stream. It also uses local cache to speed up repeated access.

Cache

To improve performance and reduce redundant API calls, the server caches downloaded files locally. The cached files are stored in the ./src/mcp_m365_filesearch/.local/downloads directory (relative to the project root). When a file is requested via get_file_content, the server first checks the cache. If the file exists in the cache, it is returned directly from the disk without making a new API call.

📄 License

This project is released under the MIT License. You are free to use, modify, and distribute the software, provided that you give appropriate credit.