Now in preview, Gemini CLI is an open-source AI agent that brings the capabilities of Google’s Gemini AI model into the developer’s terminal. Credit: Shutterstock/TippaPatt Google has introduced Gemini CLI, an open-source AI agent that brings Gemini AI capabilities right to a developer’s terminal. Unveiled June 25, Gemini CLI is a command-line AI workflow tool that connects to a developer’s tools, understands the developer’s code, and accelerates the developer’s workflow. Gemini CLI gives developers a direct path from their prompt to Google’s Gemini AI model, the company said. “Gemini CLI is an agent that runs in your terminal that brings the power of Gemini in a very lightweight way to your terminal,” said Taylor Mullen, Google senior staff software engineer, in a Google briefing on June 24. “We think Gemini CLI is going to fundamentally change how developers interact with AI.” Gemini CLI can be used for tasks ranging from content generation and problem-solving to task management and deep research. The capabilities of the multimodal reasoning model include code understanding, file manipulation, command execution, and dynamic troubleshooting, Google said. The company said it built Gemini CLI to be extensible, leveraging standards such as Model Context Protocol (MCP), system prompts, and settings for personal and team configuration. With Gemini CLI, developers can do specific tasks including the following: Query and edit large code bases in and beyond Gemini’s 1M token control window. Generate new apps from PDFs or sketches, using Gemini’s multimodal capabilities. Automate operational tasks, such as querying pull requests or handling complex releases. Ground queries using the Google Search tool built into Gemini. Use tools and MCP servers to connect new capabilities including media generation with Imagen, Veo, or Lyria. Offered via an Apache 2.0 license, Gemini CLI has been integrated with Google’s AI coding assistant, Gemini Code Assist. Users of Gemini Code Assist on free, Standard, and Enterprise plans get prompt-driven, AI-powered coding in both Visual Studio Code and Gemini CLI, Google said. Developers can use Gemini CLI free of charge by logging in with a personal Google account to get a free Gemini Code Assist license.