AionUi is a free, local, and open-source desktop application designed to serve as a unified graphical user interface (GUI) and "Cowork" platform for various command-line interface (CLI) AI tools and large language models (LLMs). It addresses common pain points associated with CLI tools, such as the inability to save conversations, single-session limitations, cumbersome file operations, and single-model support.

The core methodology of AionUi revolves around providing a comprehensive, multi-agent AI environment with a focus on usability, data security, and office automation.

Key Technical and Methodological Aspects:

1. Unified Graphical Interface for CLI Tools: AionUi automatically detects and integrates with locally installed CLI AI tools, including but not limited to Gemini CLI, Claude Code, Codex, Qwen Code, Goose AI, and Augment Code. It abstracts away the command-line interaction, providing a cohesive visual workspace. This involves an auto-detection mechanism to identify the presence of these tools and then exposing their functionalities through a user-friendly GUI.

2. Multi-Agent and Multi-Model Support: Beyond integrating existing CLI tools, AionUi supports direct integration with various mainstream LLM APIs (e.g., Gemini, OpenAI, Claude, Qwen) and local LLM deployments (e.g., Ollama, LM Studio). Users can configure custom API endpoints (e.g., http://localhost:11434/v1 for Ollama) to connect to local models. The platform allows flexible switching between different models within the same interface and supports running multiple AI agents simultaneously, each potentially utilizing a different underlying model or CLI tool.

3. Persistent Local Data Storage and Session Management: All conversation data, user configurations, and associated files are stored locally in an SQLite database. This design choice prioritizes data privacy and security, ensuring that sensitive information never leaves the user's device. AionUi supports multiple parallel chat sessions, with each session maintaining an independent context and memory. This prevents context confusion across different tasks and enhances efficiency.

4. WebUI Remote Access (7/24 AI Assistant): AionUi features a WebUI mode, enabling users to access their AI assistant from any device on the network (e.g., phone, tablet, computer) via a web browser. This functionality supports cross-device access, remote collaboration (LAN, cross-network, server deployment), and maintains local data security. Access can be secured with password management and QR code login, ensuring controlled remote interaction with the locally stored AI environment.

5. Smart File Management (AI Cowork): The platform integrates advanced file management capabilities, referred to as "AI Cowork" for files. This includes:
   • Batch Operations: One-click batch renaming and file merging.
   • Intelligent Organization: AI-driven content identification and auto-classification to organize folders.
   • File Tree Browsing and Drag-and-Drop: Intuitive file system navigation and easy import of files or folders.
   • Data Processing: AI-driven analysis and beautification of structured data, such as Excel reports.
   • Document Generation: Automated generation and formatting of various document types like PPT, Word, and Markdown.

6. Integrated Preview Panel: AionUi features a built-in preview panel that supports over 9 file formats, including PDF, Word, Excel, PPT, various code languages, Markdown, images, HTML, and Diff. This allows immediate visual inspection of AI-generated content without requiring external applications. The panel also offers real-time tracking of file changes, with synchronization between the editor and preview, and supports live editing (WYSIWYG) for formats like Markdown, code, and HTML.

7. AI Image Generation & Editing: The system integrates intelligent image generation capabilities, supporting various models (e.g., Gemini 2.5 Flash Image Preview, Nano, Banana). It also provides AI-driven image recognition and editing functionalities.

8. User Interface Customization: The interface is highly customizable, allowing users to modify colors, styles, and layout through custom CSS code, enabling a personalized user experience.

The architecture aims to provide a robust, cross-platform (macOS, Windows, Linux) solution that enhances the utility of existing AI tools by wrapping them in a feature-rich, privacy-focused environment, particularly excelling in AI office automation scenarios.