Text from a screenshot,on your Mac

Some of the text you need lives in pixels: a pasted screenshot, a locked PDF, a read-only dashboard, an image a teammate sent. On a Mac, Khint gets it out in one gesture: open the palette, drag a box, and the plain text is on your clipboard. No retyping, no Screen Recording permission, and it works on top of any app.

For product owners, business analysts, and anyone on macOS who keeps retyping text off a screen.

Cmd+Shift+4 gives you a picture,
you wanted the words

The built-in Mac screenshot saves an image: great when you want a picture, useless when you needed the text inside it. So you squint and retype: a figure off a chart you can't click, an error string in a dialog, a requirement someone photographed off a whiteboard. Khint's Capture pillar closes that last inch: it reads the region you drag and hands you the characters, ready to paste. It's one of Khint's three pillars alongside Agents and Memory, all behind the same palette.

Four keystrokes instead of four minutes

The same string, two ways to get it off the screen.

Retyping by hand
  • Squint at the screenshot
  • Retype the string, character by character
  • Second-guess a 0 versus an O
  • Switch apps to paste, lose your place
Extract text
  • Cmd+Shift+K → Extract text
  • Drag a box around the region
  • Text is on your clipboard
  • Paste it exactly where you were

Screenshot to clipboard, step by step

How the Extract text path plays out on a Mac.

  1. Open the palette from wherever you are

    Text you can't select turns up in every app: Slack, Preview, a browser tab, a remote-desktop window. Hit Cmd+Shift+Kto open Khint's palette on top of whatever you were doing. Capture lives there next to your Actions, so there's no separate app to launch. How Capture works →

  2. Pick Extract text and drag a box

    Choose Extract textand the macOS screenshot selector appears: the same crosshair you know from Cmd+Shift+4, because Khint delegates the region grab to the system's screencaptureui. Draw a box around the words you need: a figure in a locked dashboard, a paragraph in a non-selectable PDF, a label in a screenshot a stakeholder pasted into a thread.

  3. The text lands on your clipboard

    Khint reads the region and copies the plain text straight to your clipboard. A one-line overlay confirms the grab with a preview of what it got, so you stay in the doc you were writing: no window pops open, no tab switch. Paste it into a ticket, a planning doc, or a message the same way you'd paste anything else.

  4. Drop a file when it isn't on screen

    Not every image is on a monitor. A phone photo of a whiteboard, a scanned page, an exported chart: drag it into the Capturetab's drop zone or pick it from disk. PNG, JPG, WebP, HEIC and more all come back as clean text, the same as a region capture. Whiteboard photo to requirements →

  5. Feed the text into an Action, not a dead end

    Captured text uses the same paste path as a Khint Action, so it can flow straight into one: clean it up, reshape it into a user story, or summarise it for a ticket. With Atlassian or Linear connected on the integrations page, the result can even be filed as an issue from the same shortcut. Screenshot an error into a bug ticket →

Why it only asks for Accessibility

Plenty of screen tools want blanket Screen Recording access before they'll read a single pixel. Khint takes a narrower path: on macOS it delegates the region selection to screencaptureui, the OS screenshot UI, which runs with your own authority, so Khint's own process never calls a direct screen-capture API and never records your screen in the background. The one permission it does need is Accessibility, so it can paste the extracted text back where your cursor is. When you need to reason about a dense screen rather than just grab its text, Capture & ask opens a Vision chat on the same image, but the plain grab is a one-shot, clipboard-first move.

Common questions

How do I extract text from a screenshot on a Mac?

Open Khint's palette with Cmd+Shift+K and pick Extract text. You draw a box around whatever is on screen (a screenshot pasted in Slack, a read-only PDF, a number in a dashboard you can't click into). Khint reads the region and drops the plain text on your clipboard. A small overlay confirms the grab with a one-line preview, then you paste the words wherever your cursor already is. There's no separate window to switch to and no hotkey to memorise beyond the palette shortcut.

Does Khint need Screen Recording permission to do this?

No. On macOS, Khint hands the region selection to screencaptureui (the same system screenshot UI you get with Cmd+Shift+4), which runs with your own authority, so Khint's process never calls a direct screen-capture API. That means the only macOS permission it asks for is Accessibility, needed to paste the result back for you. Extracting text from a screen is not a background screen-recording capability, and Khint doesn't request one.

What if the image is a file, not something on my screen?

The Capture tab has a drop zone. Hand it a PNG, JPG, WebP, HEIC, GIF, BMP, or TIFF (or pick one from disk) and the text comes back the same way as a live region grab. So an emailed mockup, a phone photo of a whiteboard someone AirDropped you, or an exported chart all become editable text without retyping a line.

Where does my screenshot go: is the text private?

The capture is sent to Khint's vision service to read the text, then the recognised words land on your clipboard and in your local history. History bodies (the input and output text) live in a local SQLite database on your Mac and are never synced to Khint's servers. Capture also has its own daily allowance, separate from your AI Action quota, so reading screens never eats into the actions you keep for writing.

Stop retyping what's already on screen

Free with 10 AI actions and 5 captures per day. No credit card. Capture lives in the palette, one shortcut away.