Google Workspace CLI: New Command Line Interface for AI Agents & Developers

The Command Line Returns to Google Workspace

The command line interface – once a clunky, code-based method of interacting with computers – is experiencing a resurgence, driven by the rise of agentic AI. Tools like Claude Code and Kilo CLI have demonstrated the power of a scriptable interface for AI agents to execute tasks. Now, Google is bringing this approach to its Workspace suite – encompassing Gmail, Drive, Docs, Sheets, and more – with a fresh command-line interface (CLI) designed for both human developers and AI agents.

The project, dubbed googleworkspace/cli, aims to provide a unified interface for accessing Google Workspace applications and their data, eliminating the necessitate for third-party connectors. As Google Cloud director Addy Osmani announced on X, the CLI is “built for humans and agents,” covering “Google Drive, Gmail, Calendar, and every Workspace API.” The tool is available on Github as an open-source project (Apache 2.0 license).

How the Google Workspace CLI Works

For years, Google Workspace applications have been accessible via APIs, but interacting with those APIs often required building custom wrappers or relying on integration platforms like Zapier. The gws CLI offers a different approach: a single command surface with structured JSON output. In other words that commands executed through the CLI return data in a predictable, machine-readable format, making it easier for AI agents to parse and utilize the information.

Installation is straightforward using Node Package Manager (npm): npm install -g @googleworkspace/cli. The repository states that prebuilt binaries are included, and releases are too available through GitHub. A key feature is its dynamic command surface. The CLI reads Google’s Discovery Service at runtime, meaning new Workspace API methods become available without requiring updates to the CLI itself. This reduces maintenance overhead and allows the CLI to adapt to changes in the Workspace ecosystem.

Benefits for Developers and AI Agents

The CLI is designed to be useful for both direct human interaction and agent-driven workflows. For developers working in the terminal, features like per-resource aid, dry-run previews, schema inspection, and auto-pagination enhance usability. However, the real value emerges when considering AI agents. The structured JSON output, reusable commands, and built-in “skills” – over 100 are currently included for Gmail, Drive, Docs, Calendar, and Sheets – allow models to interact with Workspace data and actions without complex custom integrations.

This capability unlocks potential for automating a wide range of internal enterprise workflows. Tasks like listing Drive files, creating spreadsheets, sending Chat messages, and generating reports can be streamlined through agentic automation. The CLI provides a way to retrieve information, trigger actions, and automate repetitive processes with less bespoke code.

A Caveat: Not Officially Supported… Yet

Despite the enthusiasm surrounding the release, it’s crucial to understand that the Google Workspace CLI is not currently an officially supported Google product. The README explicitly states this, and warns users to expect breaking changes as the project moves towards version 1.0. This means enterprises should approach adoption cautiously, initially focusing on testing and evaluation in sandboxed environments.

This isn’t necessarily a sign of a lack of commitment from Google, but rather a reflection of the project’s early stage of development. It’s more akin to a promising developer tool with strong momentum than a production-ready platform for large organizations. The project’s open-source nature allows for community contributions and rapid iteration, but also introduces a degree of uncertainty.

Authentication and Existing Controls

The CLI doesn’t bypass existing Workspace access controls. Users still require a Google Cloud project for OAuth credentials and a Google account with appropriate Workspace permissions. The documentation outlines various authentication patterns for local development, continuous integration (CI), and service accounts, along with instructions for enabling APIs and resolving setup issues. This ensures that the CLI operates within the established security framework of Google Workspace.

Essentially, the CLI provides a more usable abstraction layer over existing permissions and admin controls, rather than granting unauthorized access. It simplifies the process of interacting with Workspace data, but doesn’t circumvent the underlying security measures.

MCP and the Broader Agent Interface Strategy

Some initial commentary positioned the Google Workspace CLI as a cleaner alternative to Model Context Protocol (MCP)-based setups, suggesting it could avoid the overhead of large tool definitions within the context window of language models. While there’s merit to this argument, the project itself presents a more nuanced view.

The repository includes a Gemini CLI extension that allows Gemini agents to access gws commands and Workspace agent skills after terminal authentication. It offers an MCP server mode through gws mcp, exposing Workspace APIs as structured tools for MCP-compatible clients like Claude Desktop, Gemini CLI, and VS Code. This suggests that Google isn’t choosing between CLI and MCP, but rather viewing the CLI as a foundational interface with MCP available as an option where appropriate.

What Enterprises Should Do Now

The immediate next step for enterprises isn’t widespread deployment, but targeted evaluation. Developer productivity, platform engineering, and IT automation teams should test the tool in a sandboxed Workspace environment, identifying specific use cases where a CLI-first approach could reduce integration effort. File discovery, spreadsheet updates, document generation, calendar operations, and internal reporting are all potential starting points.

Security and identity teams should prioritize a review of authentication patterns, determining how to tightly constrain permissions, scopes, and service account usage. AI platform teams should compare direct CLI execution against MCP-based approaches in real-world workflows, focusing on reliability, prompt overhead, and operational simplicity. The trend towards command-line interfaces as a common control plane for both developers and AI systems is clear, and the Google Workspace CLI represents a significant step in that direction.