More screenshots

    Works with the AI tools you already use

    Claude CodeClaude CodeCursorCursorCodex CLICodex CLIGitHub CopilotGitHub CopilotGemini CLIGemini CLI+17 more

    Mac ScreenDumps

    1

    Screenshot every virtual desktop on every Mac display in one run, named Screen 1 - Desktop 1 and so on.

    $5

    30-day refund guarantee

    Secure checkout via Stripe

    0 installsSecurity scanned

    See it in action

    You say

    I need to take a screenshot of every single Space and monitor I have open right now to document my workflow.

    Your agent does

    Terminal output from a real run on a two-display Mac - a 3440x1440 ultrawide as the main display and a 1280x1024 secondary - with 16 virtual desktops on each:

    Found 2 displays. Capture method: screencapture -R

    -- Screen 1 3440x1440 (main display) -- 16 desktops according to the system . Screen 1 - Desktop 1.png . Screen 1 - Desktop 2.png ... through Desktop 16

    -- Screen 2 1280x1024 -- 16 desktops according to the system . Screen 2 - Desktop 1.png ... through Desktop 16

    -- Check -- Screen 1: 16 images of 16 known desktops ok Screen 2: 16 images of 16 known desktops ok

    Result: 32 PNG files plus _diagnostics.txt in ScreenDumps/2026-08-21 20.26.46/, about 60 MB, produced in just under two minutes. Both displays were returned to the desktop they started on.

    What you get

    Document a full workspace layout before reinstalling macOS or handing the machine to someone elseCapture every virtual desktop across two monitors for a support ticket or bug reportProduce a visual inventory of what is open across 16 Spaces before stepping away from a projectRecord the state of every desktop before and after reorganising a multi-monitor setup

    About this skill

    The problem

    macOS prevents apps from capturing any virtual desktop (Space) that is not currently visible. Standard screenshot tools only grab the active screen, forcing you to switch to each desktop and snap it by hand, one at a time.

    What it does

    • Walks every Space on every connected display and captures each one as it becomes visible.
    • Names files by position: Screen 1 - Desktop 1.png through Screen 2 - Desktop 16.png. Full-screen apps own their own Space and are numbered separately as Fullscreen 1, 2, matching how Mission Control treats them.
    • Activates the correct display before each walk. The desktop-switch shortcut follows keyboard focus, not the pointer, which is why a naive script captures the second monitor's first desktop sixteen times.
    • Tries five capture methods at startup, including CoreGraphics directly, and keeps whichever actually writes a file. The screencapture tool behaves differently across macOS versions and can report success while producing nothing.
    • Hashes every image and signs every display with the window IDs on it. If neither changes, the desktop did not switch, so nothing is written and the display is reported as stuck instead of filling the folder with copies.
    • Returns each display to the desktop it started on, where macOS reports that reliably.

    What you get

    A single double-clickable .command file with a documented settings block at the top - output folder, image format, delays, whether to include full-screen apps - plus a SKILL.md that tells your agent how to install, run and troubleshoot it, and a symptom-by-symptom troubleshooting reference. It installs nothing, downloads nothing, and uses only tools that ship with macOS. MIT licensed.

    When something goes wrong

    Every run ends with a per-display check - Screen 1: 16 images of 16 known desktops, ok - and writes a diagnostics log recording the macOS version, which capture method won, what the others reported, and how many Spaces each display claimed. That file usually names the cause outright.

    Why this beats prompting it yourself

    macOS provides no public API for switching Spaces, so the job has to be done by driving the keyboard and verifying the result afterwards. The hard part is not taking the screenshots. It is the permission requirements, the display focus logic, and knowing when a switch silently failed.

    How to install

    Drop the file into your AI Agent. Works with Claude, Cursor, ChatGPT, and 20+ more.

    Reviews

    No reviews yet

    Be one of the first to try it. Every listed skill passes our trust checks below.

    Security scanned

    Passed our 8-point scan before listing

    Fresh listing

    Recently published to Agensi

    30-day refund

    Not a fit? Get your money back

    Trust & safety

    Security scanned

    Verified clean today

    • One-time purchase, yours forever

    Listedtoday

    Frequently Asked Questions