Quick Facts
- Google released an open-source CLI that connects AI agents to Gmail, Drive, Docs, Calendar and Sheets through one unified interface
- The tool includes over 100 pre-built agent skills and automatically updates when Google adds new API endpoints
- Built in Rust and distributed through npm, the project hit #1 on Hacker News with 571 points but remains experimental without official Google support
Google released an experimental command-line interface that gives AI agents direct access to Gmail, Google Drive, Docs, Sheets and other Workspace services through a single tool.
The Google Workspace CLI automatically builds its command structure by reading Google’s Discovery Service at runtime. When Google adds new API endpoints, the CLI picks them up without manual updates.
The tool ships with more than 100 agent skills covering common workflows across Gmail, Drive, Docs, Calendar and Sheets. Each service adds roughly 10 to 80 tools, though developers should limit selections to stay under typical client tool limits of 50 to 100 tools.
Installation requires a single npm command: npm install -g @googleworkspace/cli. The package includes pre-built binaries for different operating systems, eliminating the need for a Rust toolchain.
The CLI features a git-like workflow for editing documents. Users can pull a Google Sheet, which converts to a local folder with a .tsv file and formula.json. After local edits, users push changes back to Google Sheets.
For AI integration, the CLI supports Model Context Protocol. Running gws mcp -s drive,gmail,calendar starts an MCP server that exposes Workspace APIs as structured tools for MCP-compatible clients like Claude Desktop.
The tool includes specific integration instructions for OpenClaw, Google’s response to rising demand for AI agents that execute tasks rather than just answer questions. The documentation shows Google positioning Workspace services as “agent-ready” following OpenClaw’s success.
Enterprise developers previously relied on third-party connectors like Zapier or built custom API wrappers. The CLI eliminates this complexity with a unified interface and structured JSON output designed for both humans and AI agents.
The project hit #1 on Hacker News with 571 points, reflecting strong developer interest in AI agent tooling. However, Google labels this an experimental project under active development with expected breaking changes before version 1.0.
Security features include Google Cloud Model Armor support to mitigate prompt injection risks when feeding API data to language models. The project uses an Apache 2.0 license and remains available on GitHub.
This release signals Google’s strategy to make enterprise productivity suites more accessible to AI agents as the industry shifts toward agentic workflows that automate real business tasks.
Read more: Google Workspace CLI brings Gmail, Docs, Sheets and more into a common interface for AI agents