Google Docs MCP Server

Example Use Cases

• 1

  Search a user's Google Drive for relevant documents by query and return matching results.

• 2

  Read and extract content from Google Docs, Sheets, and Slides and receive it in agent-friendly formats (markdown, CSV, text).

• 3

  List documents within a folder or across Drive and integrate authenticated access into MCP-capable clients like Claude Desktop.

Description

This project implements a Model Context Protocol (MCP) server that enables AI assistants like Claude to search Google Drive, list documents, and read content from Google Docs, Sheets, and Slides. It handles OAuth2 authentication, stores tokens locally, and can convert document formats (Docs → Markdown, Sheets → CSV, etc.) for easier consumption by agents. The server is designed to run locally and be configured as an MCP endpoint in clients such as Claude Desktop, enabling seamless, authenticated access to a user's Drive content. Installation and developer instructions are included for running and testing the server.