
your take opens in the editor with the cut already made
the editor
Every finished recording opens in the editor. Trim, zoom, swap the backdrop, tune the narration, then export.

demos on autopilot
01 / agent
Give it a sentence. It drives your app and films the run.
02 / auto-zoom
The camera pushes in on every click, then glides back out. No keyframes.
03 / ai narration
Written from what happened on screen. Cut a clip and the voiceover follows.
04 / parallel
Queue a flow per feature. Come back to finished takes.
05 / time
Hours of recording and editing, down to one sentence.
06 / takes
UI changed? Ask for a fresh take. Every take stays saved under the video.
how it works
01
A URL and one sentence. The agent does the driving, or hit record and drive it yourself.
02
Crisp video plus every click and keystroke, so the editor knows where the action is.
03
Auto-zoom, framing, narration, music. Export the MP4 and send it.
agent-native
One install adds the CLI and an agent skill, so the agent that shipped the feature can film its demo.
terminal
pay as you go
$ curl -fsSL https://www.clankcut.com/install.sh | bash
# clankcut cli + agent skill installed
$ clankcut recordings create \
--url https://app.example.com \
--flow “sign in, create a project, export a clip”
queued · leased · take lands in the editorA Mac. Clank Cut is a macOS app, and the AI demo agent drives web apps in stock Google Chrome.
Point it at a URL and describe the flow in a sentence. The agent drives Chrome itself, clicking real buttons rather than guessing at pixels. The take lands in a project like any other recording, and if it gets stuck it stops and tells you.
It watches where you (or the agent) actually clicked and moves the camera there, settling smoothly between moments. What you preview is exactly what exports.
Yes. Trim, split, re-speed, adjust zoom regions, swap the wallpaper framing, regenerate narration, and restyle the cursor and click ripples on your own recordings.
AI narration is written from what actually happened and anchored to steps, so cutting a clip keeps the voiceover in sync. Background music generates to fit the timeline and ducks itself under speech.
Coming soon. A presenter bubble you can drop over the frame is on the way.
Yes. One command installs the clankcut CLI, the agent skill, and the desktop app if it's missing. Recordings queue as durable jobs and retries are handled for you.
A crisp MP4 that plays and scrubs smoothly wherever you drop it: Slack, Notion, a changelog, YouTube.
Pay as you go, billed per organization. No subscription tiers, no seats.
Zero recording. Real demos. macOS app for the recorder, Chrome for the agent, one editor for both.