Google has released Gemini CLI, a free and open-source AI agent that integrates Gemini into the command line interface (CLI).

Designed for developers who prefer the terminal, Gemini CLI offers a fast, direct connection to the Gemini 2.5 Pro model.

“For developers, the command line interface (CLI) isn't just a tool; it's home,” said Taylor Mullen, Senior Staff Software Engineer.

Gemini CLI supports a wide range of tasks beyond coding, including content generation, deep research, and task management.

The tool is available now in preview and supports both interactive and scripted use.

Developers Get Free Access With the Industry’s Largest Usage Limits

To access Gemini CLI at no cost, users can log in with a personal Google account and receive a free Gemini Code Assist license.

This grants access to Gemini 2.5 Pro with a 1 million token context window.

The free tier includes an unmatched usage allowance: 60 model requests per minute and 1,000 requests per day.

These limits are designed to ensure developers “rarely, if ever, hit a limit during this preview”.

Developers needing more advanced options, such as running multiple agents or choosing specific models, can use a Google AI Studio or Vertex AI key, or subscribe to Gemini Code Assist’s Standard or Enterprise plans.

Built-In Tools Enable Command Execution, Web Search, and Custom Prompts

Gemini CLI introduces tools that expand its core functionality. Developers can ground prompts with Google Search, giving the model real-time access to web content.

Built-in support for the Model Context Protocol (MCP) and bundled extensions make it extensible by design.

Users can customize the agent’s behavior through prompts and instructions tailored to their workflows.

The tool also supports non-interactive usage, allowing it to be embedded into scripts and automated systems.

Gemini CLI’s features range from file manipulation to making short videos using Veo and Imagen.

Open-Source and Extensible Under Apache 2.0 License

Gemini CLI is fully open source under the Apache 2.0 license. This transparency enables developers to inspect the code, verify security practices, and contribute to the project.

Google encourages feedback and collaboration through its GitHub repository.

The project supports emerging standards such as GEMINI.md for system prompts and settings files for both personal and team configurations.

“We know the terminal is a personal space, and everyone deserves the autonomy to make theirs unique,” said the development team.

Gemini Code Assist and CLI Share the Same Agent Technology

Gemini CLI and Gemini Code Assist now run on the same underlying technology. This integration allows developers to switch between the terminal and IDE environments seamlessly.

In Visual Studio Code, Code Assist supports agent mode, where developers can prompt the assistant to write tests, fix bugs, and build features.

“Code Assist will relentlessly work on your behalf to write tests, fix errors, build out features or even migrate your code,” said Ryan J. Salva, Senior Director of Product Management.

Gemini Code Assist’s agent builds multi-step plans, recovers from failed implementations, and suggests alternative solutions.

Agent mode is available in the Insiders channel at no additional cost for free, Standard, and Enterprise users.

PHOTO: GOOGLE

This article was created with AI assistance.