Gemini CLI puts AI directly in your terminal
Google has officially launched CLI, a powerful AI agent designed to revolutionise how developers interact with their command line. This open-source tool brings the intelligence of Gemini 2.5 Pro straight into your terminal, enabling natural language coding, file manipulation, and much more — all from a lightweight, local interface. The release went live globally on 25 June and is already making waves across the development community.
Gemini CLI: A smarter terminal for developers
The CLI combines Google’s most advanced AI model with the familiar command line interface (CLI) that developers rely on daily. Traditionally, the CLI requires users to memorise cryptic commands and syntax, which can be time-consuming and error-prone. But CLI changes the game.
“You can tell your computer what you want to achieve rather than struggling with how to say it in the computer’s native tone,” said Taylor Mullen, senior staff software engineer at Google. He added that Gemini CLI eliminates the need to memorise obscure terminal commands, giving developers a smoother, more intuitive experience.
Integrated with Gemini Code Assist
The new CLI seamlessly integrates with its Code Assist, Google’s AI-powered coding assistant available in VS Code. Whether you’re a student, hobbyist or enterprise developer, Gemini CLI offers consistent support across environments.
This shared technology allows developers to:
- Write and explain code in plain English
- Debug and fix errors
- Automate tasks and integrate with scripts
- Ground prompts using real-time Google Search results
- Customise prompts and workflows using Model Context Protocol (MCP)
The integration means that developers using of Code Assist can access the same capabilities through the terminal, bringing flexibility and power wherever they work.
Unmatched usage limits for all users
Google has set the bar high by offering generous free-tier limits for users of CLI. Anyone with a personal Google account can access the tool and receive a free Code Assist license, which includes:
- 60 model requests per minute
- 1,000 model requests per day
- Access to Gemini 2.5 Pro
- A 1 million token context window
Professional developers who need more advanced options can use usage-based billing through Google AI Studio or Vertex AI, or opt for Standard and Enterprise licenses.
Gemini CLI supports deep research and content generation
While coding remains the core strength of this CLI, it supports a wide range of other tasks. From problem solving and task management to content generation and research, developers can ask Gemini CLI to help with virtually any terminal-based activity.
In fact, Google cites use cases such as creating short videos, conducting deep research, managing projects, and even troubleshooting system-level issues — all using natural-language prompts.
Open-source and community-driven
Google released Its CLI under the Apache 2.0 open-source license, encouraging developers to participate in its evolution. Developers can inspect the code, propose new features, and report bugs directly through the GitHub repository: github.com/google-gemini/gemini-cli.
The tool supports full customisation for individuals and teams, building on standards like system prompts and MCP. Developers can personalise how the tool behaves and interacts with their terminal — a key advantage in professional environments.
Gemini CLI supports security and flexibility
To ensure safety and transparency, Its CLI prompts users for confirmation before executing sensitive commands. It also uses layered sandboxing for additional security, particularly when operating in local environments.
Ryan J Salva, senior director of product management, highlighted that Google’s APAC data centres serve developers in India and the broader Asia-Pacific region, underlining the company’s commitment to global accessibility.
Also Read: UK Climate Goals Within Reach, But Cheaper Electricity Key to Clean Tech Shift
Getting started with Gemini CLI
Getting started with Its CLI takes less than a minute. Developers only need an email address to activate the free license. The setup is quick, and the documentation provides clear guidance for installing the tool on Mac, Windows or Linux.
It CLI represents a bold step toward more intelligent and intuitive development environments. By bringing AI directly into the terminal, Google empowers developers to do more with less — and work smarter, not harder.
