OpenViSTAI · Agents · Blockchain · Crypto
Menu
MCP & Tooling

Google Workspace MCP Server

Google Workspace MCP Server is a Python MCP server and CLI that exposes Google Workspace services to AI assistants and developer tools. The repository describes over 120 tools across 12 services, including Gmail, Calendar, Drive, Docs, Sheets, Slides, Forms, Tasks, Contacts and Chat, plus Search. It supports tool tiers, local stdio and streamable HTTP transports, and multiple documented authentication and deployment

Independent OpenViST project profile, based on linked sources.

Google Workspace MCP Server — OpenViST project profile in MCP & Tooling.
OpenViST project illustration. See the linked sources for product details.
Understand the project

Practical overview

When it fits

Use it when an MCP-capable AI client or developer workflow needs access to Google Workspace through a server or CLI, and you can configure the necessary Google OAuth credentials and APIs.

Who it is for

  • Developers connecting AI assistants or MCP clients to Google Workspace.
  • Individuals who want a CLI for Workspace tasks.
  • Teams evaluating a self-hosted connector with documented multi-user authentication and deployment options.

Documented uses

  • Connect an AI client to Workspace services through MCP.
  • Choose a tool tier or select particular services for an integration.
  • Use a CLI to interact with Workspace tasks.
  • Configure an HTTP-based or local server deployment.
Before you choose

Requirements & limits

The repository is public and displays an MIT license. The README indicates Python 3.10 or newer and describes a self-hosted setup using a Google Cloud OAuth client and enabled APIs.

  • Setup depends on creating an OAuth client and enabling the Google APIs for the chosen services.
  • The README documents both stdio and streamable HTTP transports; its OAuth 2.1 example requires HTTP.
  • Tool tiers and service selection offer different scopes of exposed functionality.
  • Remote or shared HTTP deployment has additional authentication considerations, including OAuth 2.1 multi-user auth or an external auth mode.

Check current requirements in the official documentation. Source-based guidance does not mean OpenViST has run or independently tested this project.

Source-based practical guide

Prepare and connect a local Google Workspace MCP server

Configure OAuth and Google APIs, launch the server over HTTP, and add its endpoint to Claude Code or VS Code.

Before you start

  • Python 3.10+ and uv
  • An OAuth client created in Google Cloud Console
  • Google APIs enabled in the same project for the services you plan to use
  • A supported MCP client such as Claude Code or VS Code

Steps

  1. Enable the Google APIs for the services you intend to use, then export GOOGLE_OAUTH_CLIENT_ID and GOOGLE_OAUTH_CLIENT_SECRET.

  2. Launch the server with uvx workspace-mcp --transport streamable-http --tool-tier core.

  3. Connect Claude Code with claude mcp add --transport http workspacemcp http://localhost:8000/mcp, or add an HTTP server entry with that URL to VS Code's .vscode/mcp.json or user mcp.json.

  4. Complete browser OAuth on the first authenticated call and confirm the available tool list matches the selected tier.

Expected outcome

The local server is configured for the core tool tier and its HTTP endpoint is registered with the chosen client.

Guide limitations

  • The server listens at http://localhost:8000/mcp by default.
  • The recipe uses the core tier; extended and complete tiers expose additional tools.
  • OAuth 2.1 multi-user mode requires HTTP transport and cannot be combined with --single-user.
  • Local HTTP development uses OAUTHLIB_INSECURE_TRANSPORT=1; this setup does not document production deployment.
Evidence and transparency

Sources & review

Source record details
Source availability
Active
Sources checked
9 October 2026
Original website title
taylorwilsdon/google_workspace_mcp · Official homepage title · retained in its source language