desktop▌

Vision-driven desktop automation for native apps using natural language commands and screenshots. \n \n Controls macOS, Windows, and Linux desktops entirely from visual input; no DOM or accessibility labels required \n Operates synchronously with a screenshot-analyze-act loop: connect, observe screen state, execute high-level actions via natural language prompts, then disconnect \n Requires a vision-capable AI model (Gemini, Qwen, Doubao, or similar) configured via environment variables; support

Electron desktop development guide for LobeHub's main-renderer architecture. \n \n Covers controller creation, IPC type definitions, renderer services, and store actions for adding new desktop features \n Includes security best practices: input validation, limited API exposure, and preload script patterns for safe main-renderer communication \n Provides structured references for feature implementation, local tools workflow, menu configuration, and window management \n Emphasizes async methods, b

Rust has emerged as a premier language for building desktop applications that combine native performance with memory safety. The ecosystem offers two main approaches: Tauri for hybrid web UI + Rust backend apps (think Electron but 10x smaller and faster), and native GUI frameworks like egui, iced, and slint for pure Rust interfaces.

LobeHub desktop is built on Electron with main-renderer architecture:

Control mouse, keyboard, and screen for cross-platform desktop automation. \n \n Five command categories: mouse control (movement, clicks, drag, scroll), keyboard input (typing, hotkeys, key presses), screen capture and analysis (screenshots, image/text location via OCR), message dialogs (alerts, confirmations, prompts), and application control (open, focus, list windows) \n Supports Windows, macOS, and Linux with platform-specific shortcuts and application launching \n Image location with confi