Gemini-CLI-UI  by cruzyjapan

Web UI for Google's Gemini CLI

created 1 month ago
307 stars

Top 87.1% on SourcePulse

GitHubView on GitHub
Project Summary

This project provides a responsive, web-based user interface for Google's Gemini CLI, enabling AI-assisted coding from any device. It targets developers who want a more intuitive and accessible way to interact with Gemini CLI's features like chat, terminal access, file editing, and Git integration, offering a unified experience across desktop and mobile.

How It Works

The UI employs a modern web architecture with a React/Vite frontend and a Node.js/Express backend. Communication between the frontend and backend, including real-time chat, is handled via WebSockets. The backend manages Gemini CLI process spawning, session persistence using JSONL files, and file system access. An integrated SQLite database handles user authentication and session management, ensuring secure access and data storage.

Quick Start & Requirements

  • Installation: Clone the repository, run npm install, copy .env.example to .env and configure, then start with npm run dev.
  • Prerequisites: Node.js v20+, Gemini CLI installed and configured.
  • Links: Repository

Highlighted Details

  • Responsive design for seamless use on desktop, tablet, and mobile.
  • Integrated file explorer with live editing and syntax highlighting.
  • Git integration for viewing, staging, and committing changes.
  • Session management for resuming conversations and tracking history.
  • Model selection for choosing different Gemini models.

Maintenance & Community

The project is based on Claude Code UI and has undergone significant adaptations for Gemini CLI. Updates and new releases can be followed by starring the repository.

Licensing & Compatibility

Licensed under GNU General Public License v3.0 (GPL v3.0). This license is copyleft, meaning derivative works must also be distributed under GPL v3.0. Compatibility with closed-source projects may be restricted due to these terms.

Limitations & Caveats

Gemini CLI tools are disabled by default and require explicit enabling in settings for security. The "YOLO Mode" bypasses confirmation prompts and should be used with caution. Ensure Gemini CLI is initialized in at least one project directory for the UI to discover projects.

Health Check
Last commit

2 weeks ago

Responsiveness

Inactive

Pull Requests (30d)
6
Issues (30d)
0
Star History
309 stars in the last 30 days

Explore Similar Projects

Starred by Koray Kavukcuoglu Koray Kavukcuoglu(Chief AI Architect at Google; CTO of Google DeepMind), John Resig John Resig(Author of jQuery; Chief Software Architect at Khan Academy), and
20 more.

gemini-cli by google-gemini

2.1%
70k
AI agent for terminal workflows
created 4 months ago
updated 3 hours ago
Feedback? Help us improve.