@quanru

midscene-computer-browser automates web interactions via vision-driven screenshots, supporting headless puppeteer, CDP remote debugging, and chrome extension bridge modes. executes click, type, scroll, and form interactions with AI-grounded element targeting and assertion-based validation.

midscene-android-automation provides vision-driven ui control for android devices via screenshot analysis and natural language commands. no dom or accessibility labels required. supports taps, swipes, text input, app launches, and assertions through a synchronous screenshot-analyze-act loop.

midscene-computer-automation provides vision-driven desktop control via natural language, operating on screenshot analysis without dom dependencies. covers macos, windows, linux with synchronous cli commands for click, type, drag interactions on native and electron apps.

Vision-driven iOS device automation using Midscene CLI. Operates entirely from screenshots — no DOM or accessibility labels required. Can interact with all v...

Vision-driven browser automation using Midscene Bridge mode. Operates entirely from screenshots — no DOM or accessibility labels required. Can interact with...

Vision-driven HarmonyOS NEXT device automation using Midscene. Operates entirely from screenshots — no DOM or accessibility labels required. Can interact wit...