openai/screenshot

screenshot is an agent skill from OpenAI's curated skills collection that captures screenshots at the operating-system level on macOS, Linux and Windows. It supports the full screen, a specific app or window, a pixel region and multi-display setups, and it uses direct OS commands as a fallback when the helper scripts are not available. It is meant for when you explicitly ask for a desktop or system screenshot, or when a tool-specific capture is not suitable. The bundled scripts choose the right capture tool, work out the save path and run permission checks, so you do not have to remember a different command on each system or deal with the macOS permission prompt each time. Typical uses include capturing a specific app window by name on macOS for visual inspection, taking a pixel-region screenshot to isolate a UI element while debugging, saving a full-screen capture to the default location on request, running a preflight permission check before capturing on macOS to avoid repeated prompts, and capturing the active window on Linux with scrot or gnome-screenshot. It suits developers and support staff who need a quick, reliable capture of what is on screen to share or to inspect. Screenshots can show private information on the screen, so check the image before you share it. The skill is listed on officialskills.sh and installs from the openai/skills repository. Install it with npx skills add https://github.com/openai/skills --skill screenshot.…