iOfficeAI has released AionUi v2.1.4, a free, open source desktop application that wraps 20+ AI agent command-line tools in a single graphical workspace. The app is available for Windows, macOS, and Linux under the Apache-2.0 license.
AionUi auto-detects installed agent CLIs on the user’s machine and presents them through a unified interface. Supported agents include Claude Code, Codex, OpenClaw, Hermes Agent, Qwen Code, Goose AI, GitHub Copilot, Cursor Agent, OpenCode, Gemini CLI, and others, according to the project’s GitHub repository and its listing on Hermes Atlas.
Built-in Agent, No CLI Required
The app ships with its own agent engine that works immediately after installation. Users sign in with a Google account or paste an API key from any supported provider (OpenAI, Anthropic, Gemini, xAI, OpenRouter, AWS Bedrock, or local models via Ollama and LM Studio). No separate CLI tools need to be installed for the built-in agent to function.
According to Firethering’s review, the built-in agent has full file read/write access, web search capabilities, image generation, and MCP (Model Context Protocol) tool support. The app also includes 21 pre-built assistants for specific workflows: PPT creation with Morph animations, Word document generation, Excel report building, pitch deck creation, and academic paper writing.
Multi-Agent Coworking and Automation
AionUi’s core differentiator from single-agent GUIs is simultaneous multi-agent operation. Users can run Claude Code on a coding task, OpenClaw on system automation, and Hermes Agent on research in parallel sessions within the same workspace.
The app supports 24/7 unattended automation through a built-in cron scheduler. Users can configure recurring AI tasks that execute on schedule without manual intervention. Remote access is available through WebUI, Telegram, Lark, DingTalk, and WeChat, letting users monitor and interact with their agent workspace from mobile devices.
The Consolidation Pattern
AionUi reflects a growing trend in the agent ecosystem: as the number of specialized CLI agents proliferates, developers increasingly need a single interface to manage them. The app’s auto-detection feature scans the local machine for installed agent binaries and surfaces them in a unified panel, eliminating the need to switch between terminal windows.
The project is open source on GitHub with documentation available in nine languages. The app requires a minimum of 4GB RAM and 500MB of disk space.