Google AI releases gws, a command line tool for Workspace APIs

In short: Google AI has released gws, an open-source command line tool that makes it easier to connect Google Workspace services to apps and AI agents, but it is not an official Google product.

What happened

Google AI published gws, a free, open-source tool you run from a computer’s command line (a text-based way to give your computer instructions). It is designed to work with Google Workspace services like Drive, Gmail, Calendar, and Sheets.

The main idea is to reduce the repetitive setup work developers usually have to write when using Google Workspace APIs. An API is like a waiter that takes your app’s request to a service and brings back the result. gws provides one consistent set of commands across services and returns results in structured JSON (a common, organized text format that machines can read easily).

Google AI says gws can also help AI agents connect to Workspace. It supports MCP, short for Model Context Protocol, which is a standard way for AI tools to plug into outside services (like using the same type of adapter for different devices). It can run as an MCP server, includes more than 100 pre-built “Agent Skills” for common tasks, and can use Model Armor to help filter risky instructions, such as prompt injection (when hidden text tries to trick an AI into doing the wrong thing).

Google notes this is community-driven and not yet an official Google product. This sits alongside Google’s official Workspace Studio, a no-code tool that started rolling out more broadly in early March 2026.

Why it matters

For teams trying to let an AI assistant do useful work in Workspace, like finding files or drafting emails based on calendar context, gws could make connections faster and less error-prone. The catch is that it may change more quickly and have fewer guarantees than official Google products.

Source: Arstechnica