AionUi provides a free, local, open-source GUI application for the Gemini CLI, enhancing the chat experience for developers with features like multi-tasking, code diff views, and file management. It aims to bridge the gap between powerful AI capabilities and everyday usability by offering a modern, efficient interface for complex AI workflows.

How It Works

AionUi is built using Electron for its desktop application framework, with a React frontend utilizing the Arco Design UI component library and UnoCSS for styling. It integrates with the Google Gemini API via its CLI core, supporting multiple authentication methods including direct API keys, Vertex AI, and OAuth. The application features a persistent local chat history, a visual file tree for project management, and a code diff view for comparing file changes, all within a natural chat UI.

Quick Start & Requirements

• Install: Clone the repository (git clone https://github.com/office-sec/AionUi.git), navigate into the directory (cd AionUi), install dependencies (npm install), configure your Google Gemini API key in the application settings, and start the application (npm start).
• Requirements: Node.js >= 16.0.0, npm >= 8.0.0, and a Google Gemini API Key.
• Build: Commands are provided for building macOS and Windows versions.
• Docs: Official Site

Highlighted Details

• Enhanced Chat Experience with multi-conversation support and persistent local history.
• File & Project Management includes a visual file tree, drag-and-drop file upload, and a code diff view.
• Developer Workflow features include full Gemini API integration and rich Markdown rendering.
• Future vision includes evolving into a Universal AI Agent Platform supporting multiple LLMs and agent types.

Maintenance & Community

The project welcomes issues and pull requests. Links to report bugs and request features are provided.

Licensing & Compatibility

This project is licensed under the Apache-2.0 License, which is permissive and generally compatible with commercial and closed-source applications.

Limitations & Caveats

The project is currently focused on the Gemini CLI, with plans to expand to other terminal-based and browser agents. While the roadmap indicates multi-LLM support, current implementation details are not specified.