Google Opens Up Workspace for AI Agents Like OpenClaw

Google is releasing a new open-source command-line interface (CLI) for Google Workspace, designed to make it much easier for both developers and autonomous AI assistants to interact with services like Gmail, Google Drive, and Google Calendar (via PC World).

The new tool is simply called “gws.” According to the official documentation, it is “built for humans and AI agents.” While many tech tools focus on a pretty interface for people to click through, Google has made sure this one speaks the language of modern AI models.

One of the most interesting parts of this release is how it works with a viral open-source project called OpenClaw. If you haven’t heard of it yet, OpenClaw is an assistant that runs on your own computer rather than in the cloud. Its creator, Peter Steinberger, famously described it as “AI that actually does things.” By connecting OpenClaw to Google’s new CLI, users can let their AI assistant clear out their inbox, draft replies, or even find specific documents in a massive Drive folder.

Technically, the CLI is quite clever. Instead of having a fixed list of things it can do, it reads Google’s own “Discovery Service” whenever you run it. This means that as soon as Google adds a new feature or a new way to use an app, the tool updates itself automatically. For those who want to get technical, it also includes over 40 “Agent Skills” and works as a Model Context Protocol (MCP) server.

You can list files, create spreadsheets, and even send messages in Google Chat using simple text commands. Because the output is in a clean format, an AI agent can “read” the results and decide what to do next.

The project is currently hosted on GitHub, and because it is open source, anyone can contribute to its growth. It is worth noting that while this is a Google-led project, the company states it is not an “officially supported” product in the traditional sense.