Gemini CLI is a terminal-first AI agent that brings Gemini models into the command line for codebase querying, code generation, and workflow automation. It runs locally via Node.js but depends on the remote Google Gemini API for inference, requiring authentication and network access.
Project overview
It integrates file operations, shell commands, web fetching, and multimodal inputs directly into the terminal, along with scriptable automation and Model Context Protocol support for custom integrations.
Project type
MCP · AI Agent · AI Coding
Use cases
Documents & Office · Coding & Development
Deployment
Refer to project documentation
License
Apache-2.0
Best for
Developers who work primarily in the command line and want to automate operational tasks, query local files, and generate code using Gemini models.
Key capabilities
Query and edit large codebases, generate new apps from PDFs, images, or sketches using multimodal capabilities, and debug issues using natural language.
Automate operational tasks like querying pull requests or handling complex rebases, and run non-interactively in scripts for workflow automation.
File operations, shell commands, and web fetching are integrated directly into the CLI.
Extensible support for custom integrations using MCP servers.
Save and resume complex sessions via checkpointing.
Tailor behavior for your projects using custom context files.
Limitations and risks
Inference requires the external Google Gemini API, meaning data is sent outside the local environment for processing.
Preview and nightly releases may contain regressions or other outstanding issues.
Getting started
Run instantly with npx without prior installation using npx, or install globally via npm. Authenticate using a Google Account or API Key, then execute the 'gemini' command in the terminal.
Alternatives and comparisons
Provides an out-of-the-box multi-agent orchestration harness with automated model routing and context injection.
A native desktop and CLI AI agent that works with any LLM and connects to numerous extensions.
GitHub project description: An open-source AI agent that brings the power of Gemini directly into your terminal.
README: 🔧 Built-in tools**: Google Search grounding, file operations, shell commands, web fetching. - **🔌 Extensible**: MCP (Model Context Protocol) support for custom integrations. - **💻…
README: New preview releases will be published each week at UTC 23:59 on Tuesdays. These releases will not have been fully vetted and may contain regressions or other outstanding issues.
README: New stable releases will be published each week at UTC 20:00 on Tuesdays, this will be the full promotion of last week's `preview` release + any bug fixes and validations.
README: - Query and edit large codebases - Generate new apps from PDFs, images, or sketches using multimodal capabilities - Debug issues and troubleshoot with natural language