Three kinds of capture. One delivery path.
The whole screen, a dragged region or a single window. Start one from the board, the tray or a hotkey and it arrives the same way every time.
For Windows today. macOS and Linux are coming soon.
Capture
Drag a box.
That’s the tutorial.
Try it on the right. The whole screen means your primary display, edge to edge. A window means that window, not the rectangle it sits in. A region is a box you drag, across every display, with Esc to back out and a drag under four pixels read as a click.
- Drag across every display
- 3s, 5s or 10s delay
- Six global hotkeys
Drag to select a region · Esc to cancel
Region: 880 × 600 pixels of a 1920 × 1080 display.
Drag on the picture. With it focused, arrow keys move the selection, shift with an arrow key resizes it, and Esc cancels.
This is the overlay running in the page. The scrim, the four pixel rule and Esc behave as the app does; the size readout and the moving dashes are drawn for this demo, and the app’s own overlay prints no dimensions. In the app the window list carries each window’s title over the app it belongs to, and the overlay covers your whole desktop rather than a box on a page. Nothing here is captured, saved or sent anywhere.
Nothing here asks you to shoot the screen and crop it after
The whole screen, a dragged region or a single window, one at a time, and the app makes all three the same size of job.
Whole screen
One display, edge to edge. It is the shot to take when you know the page is wrong and you do not yet know which part of it.
A dragged region
Drag a rectangle over anything on screen. Escape cancels, and a drag under four pixels is read as a click and cancels too, so a stray press never hands you an empty picture.
A single window
Pick from a list of what is actually open, each row showing the window title over the name of the app it belongs to. You get that window, not the rectangle it happens to sit in.
Three ways in, and they cannot disagree
You can start a capture from the three buttons on the board, from the tray menu, or from a global hotkey while another app has focus. All three funnel into one function in the Rust core, and every capture reaches the board as a single event.
That is one place where what happens next is decided, rather than three. A shot taken from the keyboard cannot quietly do something different from a shot taken from the board, because there is no second code path for it to do it in.
The delay is on the buttons, and only on the buttons
The board carries a delay selector: none, three, five or ten seconds. It belongs to the three capture buttons, because that is where you set a shot up before you take it. Press the button, open the menu you need in the picture, and let the countdown run.
The tray and the global hotkeys always shoot instantly, by design. A hotkey that silently waits reads as a failed hotkey, so those two paths never take the delay.
Copy it, and it is on the board before you paste
One setting decides what happens the moment a capture lands, and it applies to the buttons, the tray and the hotkeys alike.
- Open the editor
- The default. The annotation editor opens on the capture straight away, which is the mode to leave on if most of your shots get an arrow or a redaction before they go anywhere.
- Save straight to the folder
- Written to your capture folder with no questions asked. With this off, saving asks where with a normal Save As dialog, so a file never lands somewhere you did not choose.
- Copy to the clipboard
- Straight into the next chat window or ticket, while the board composer takes the same picture.
Images that never went through a capture
The composer takes an image three other ways. Paste one with Ctrl+V, drag a file onto the drop zone, or press Add image and pick one off the disk. From there it is a capture like any other: annotate it, note it, save the card.
Add image opens a local file picker and nothing else. The button was called Upload once and the label was changed, because it uploads nothing and a name that advertises a cloud step the app does not take is a name that has to be corrected in support instead.
In the board itself, S takes a capture and A opens the editor once an image is staged. Both stand down while you are typing in a field or a dialog is open. Ctrl+Enter saves the card from the note field.
Six global shortcuts, and the app names the one it actually bound
Six actions reach the app from anywhere on the machine. Every one of them is rebindable in Settings, captured by pressing the combination you want rather than typed as text.
- Capture screen
- Ctrl+Shift+S
- Capture region
- Alt+Shift+R
- Capture window
- Alt+Shift+W
- Start or stop recording
- Alt+Shift+E
- Pause or resume
- Alt+Shift+P
- Copy the last card
- Alt+Shift+C
A modifier is required. A bare key would be taken out of every other app on the machine for as long as SnapBuddy is running, which is a trade nobody agreed to when they set a shortcut.
If another app already owns a combination, the app moves down a list of alternatives for that action and tells you which one it used, naming both. Every action carries its own list, so two of them can never quietly land on the same key.
Every card is a picture and a markdown file beside it
Both land in the folder you choose. The installer is signed and comes from the same releases repo the app checks for its own updates. Windows today, and macOS and Linux are coming soon.