Skip to main content

NEWExport faster

OpenScreenA free, open-source screen recorder and video editor

Native capture, local AI, no paywall.

DownloadRead the docs

Scroll down

Skip the editor — go to downloads

Five things you will actually do​

Style

Swap the background

Image, color or gradient behind your recording — no re-shoot.

Effects

Frame it your way

Padding, motion blur, shadow, roundness — every effect composites live.

Cursor

A cursor worth watching

Size, smoothing, motion blur, click bounce — every move reads on screen.

Timeline

One click, every zoom placed

Zooms, speed ramps, trims, comments — each edit lands as a pill on the timeline.

Transcript

Edit video like text

Delete a word or a silence; the cut lands on the timeline. Nothing destructive.

Recorder, captions, agent, encoder.​

record

It records with the operating system, not around it.

Pick a window or a display. macOS goes through ScreenCaptureKit and Windows through Windows Graphics Capture — the same capture path the system uses itself. The pointer is recorded as data rather than burned into the pixels, which is the only reason you could restyle it further up this page.

ScreenCaptureKit on macOS · Windows Graphics Capture · system audio without an extra driver

export

Then it writes the file.

MP4 from 720p up to source, at 24, 30 or 60, in H.264 or H.265 — or a GIF. The encode runs on your machine and counts frames while it does. No queue, no account, no watermark, and the file is on disk when the bar fills.

H.264 / H.265 · 24, 30, 60 fps · no watermark

captions

Transcription runs on your machine.

whisper.cpp ships with the app, and the model downloads once on first use — after that it works with the network off. The audio never leaves the laptop, and what comes back is editable text: set the typeface, the size, the colour and the position, then burn it into the render.

whisper.cpp · 99 languages · offline after first run

agent

Or say which parts to cut.

The wizard further up this page places zooms by watching where your cursor went. The agent goes further: it reads the actual transcript and the actual timeline, so it answers with timecodes you can go and check — which spans it will cut, and how much that saves. Every edit it makes is an ordinary undoable one, and it needs a provider key you supply. Nothing runs until you connect one.

bring your own key · off by default · every edit is undoable

Also true

Three things a screenshot can't show.

MIT, free forever

No paywalls, no premium tier, no usage caps. Every feature ships free for personal and commercial use.

Nothing is uploaded

Recording, transcription and rendering all happen on your machine, and your video never leaves it. Text leaves only when you ask: the chat panel and caption translation, each with a key you supply. Transcription downloads its 264 MB Whisper model once, on first run.

Windows, macOS, Linux

One source tree, native capture on each. A .dmg, an .exe, a .deb, a .rpm, a .pacman, an AppImage and a Nix flake.

Quick start

Download and install

macOS.dmg
# drag OpenScreen to Applications, then
xattr -rd com.apple.quarantine /Applications/Openscreen.app

ScreenCaptureKit native capture, real cursor + click effects, native webcam.

Windows.exe
# run the installer
Nothing to type — double-click and go.

Windows Graphics Capture, system audio out of the box, native webcam.

Linux.deb
# download the .deb from Releases, then
sudo apt install ./Openscreen-Linux-*.deb

Browser-pipeline capture; needs PipeWire for system audio.

The macOS line is only needed if Gatekeeper blocks the app. Linux also ships .rpm, .pacman, an AppImage, and a Nix flake — every artifact is on the Releases page, and Installation has the full steps.