Google has introduced a new AI agent called Gemini CLI that directly works inside the terminal. Just like Anthropic’s popular Claude Code terminal agent, the Gemini CLI agent can work with your codebase, debug issues, and streamline your workflow from the command line interface.
Gemini CLI can perform Google searches to find the latest information, tap MCP servers, write code, and automate tasks within your existing workflow. On top of that, Gemini CLI is fully open-source under the Apache 2.0 license, and developers can contribute to the project.
Besides that, Google has also integrated Gemini CLI into Gemini Code Assist which you can use in VS Code. If you want to work in an IDE, Google has ensured that you can leverage all the functionalities of Gemini CLI in VS Code.
The best part is that Google is offering Gemini CLI free of charge. Yes, you read that right. You get access to the flagship Gemini 2.5 Pro model and a context window of 1 million tokens. Not only that, you can process 60 model requests per minute and 1,000 requests per day without paying anything. That’s so much better than Claude Code.
You can also define the Gemini.md file to guide the agent and add your coding preferences. With the rise of coding agents such as OpenAI’s Codex and Anthropic’s Claude Code, Google has also entered the race with this terminal-based AI agent.
Source link