An MCP server implementation that provides advanced file search within Microsoft 365, enabling discovery, metadata analysis, and content access from SharePoint and OneDrive. It exposes two core tools: search_m365_files and get_file_content. The search tool accepts a query string and returns an array of file metadata including content summary, drive ID, and file ID, enabling precise discovery. The get_file_content tool retrieves file content as a binary stream for a given drive ID and file ID, and it uses a local cache to speed up repeat access. Cached files are stored in ./src/mcp_m365_filesearch/.local/downloads, and when a file is requested, the server first checks the cache and serves from disk if available to reduce API calls and latency. The server is designed to integrate with business workflows by exposing content and metadata from SharePoint/OneDrive to downstream processes. The README also provides usage guidance for Claude Desktop with a JSON config, and Microsoft Graph app registration steps: register an app, note Client ID and Tenant ID, create a Client Secret, grant Sites.Read.All and Files.Read.All permissions, and grant admin consent. MIT license.