Google has officially launched Gemini CLI, a powerful open-source AI tool that brings the intelligence of Gemini right into your terminal. This CLI tool is designed for developers and gives them direct access to Gemini’s advanced capabilities, right from the command line.
What Is Gemini CLI?
Gemini CLI is a command-line interface version of Google’s Gemini AI. While it’s great at coding tasks, it is not limited to just that. You can use it to generate content, solve problems, summarize documents, and even automate your workflows. It works locally and runs directly in your terminal, so you do not need a heavy IDE to get started.
Gemini CLI also integrates with Gemini Code Assist, which is Google’s AI coding assistant for VS Code and other development tools. This means whether you prefer a terminal or a full-fledged IDE, the Gemini agent has you covered.
Gemini CLI is completely open source under the Apache 2.0 license. It means you can inspect and modify the code or build and contribute your own extensions. Developers are encouraged to contribute via the project’s GitHub repo to help shape the future of this AI agent together.
What Can Gemini CLI Do?
Gemini CLI is packed with features for modern developers:
- Write and debug code in natural language
- Summarize or analyze files
- Execute commands based on your prompts
- Use Google Search grounding to provide real-time web context
- Customize prompts and behavior with settings
- Run non-interactively inside scripts for automation
It also supports the Model Context Protocol (MCP) and bundled extensions, making it highly extensible and customizable for personal or team use.
How to Use Gemini CLI
Make sure you have Node.js v18+ installed. Then use the following command.
npm install -g @google/gemini-cli
geminiWhen prompted, sign in with your Google account to activate your free Gemini Code Assist license. The license grants you up to 60 model requests per minute and 1,000 model requests per day using Gemini 2.5 Pro.
If you want more control or higher request capacity, you can generate an API key from Google AI Studio and Set it as an environment variable:
export GEMINI_API_KEY="YOUR_API_KEY"
