Narrated App Tour From a Screen Recording
A narrated vertical tour of your app from one screen recording.
- Pipeline
- Motion graphics (vector)
- Format
- 9:16
- Upload
- A screen recording of your app or site, landscape is fine.
The prompt
Type your own words into the fields: the [BLANKS] fill in as you go. Copy it and send it to Claude or ChatGPT with Cut Express connected.
Prompt
Make a [LENGTH IN SECONDS]-second 9:16 narrated tour of [YOUR PRODUCT] from the screen recording I uploaded. Inspect it first (contact sheets) and pick the moments where [KEY SCREENS] are on screen. Use fit blur so the landscape recording sits in the vertical frame, with slow punch-ins on the part the voice talks about. Voice-over: make it with audio, [VOICE], this exact script: "[SCRIPT]" Captions from the voice track: bold preset, highlight [CAPTION COLOR], bottom third, 2 lines at most. A whoosh on each section change (built-in sound effects). [MUSIC] under the voice, ducked. End card for the last 2.5 seconds on [BRAND COLORS]: the logo or the name, and "[CALL TO ACTION]". Show me the timeline before you render.
The blanks
- [LENGTH IN SECONDS]
- How long the finished video runs, as a number of seconds. Reels and Shorts hold best between 15 and 60.
- [YOUR PRODUCT]
- The product's name and what it does in one line, for example "Lumen, a budgeting app for freelancers".
- [KEY SCREENS]
- The screens to show, in order, for example "the dashboard, the invoice editor and the payment page".
- [VOICE]
- The voice-over you want, for example "a warm female voice, English, unhurried" or "a deep male voice, Spanish".
- [SCRIPT]
- The exact words of the voice-over. About 2.5 words per second of video, so 50 words for 20 seconds.
- [CAPTION COLOR]
- The color of the highlighted word in the captions, as a name or a hex code like #FFD23F.
- [MUSIC]
- A track you uploaded (say its file name), or a mood and genre for the stock music search, like "upbeat lo-fi" or "cinematic, builds slowly".
- [BRAND COLORS]
- Your colors as hex codes with their roles, for example "background #0B0B0A, accent #FFD23F".
- [CALL TO ACTION]
- What the viewer should do at the end, in the words to show, for example "Try it free at lumen.app".
Tips
- Write the script yourself, or ask your AI for three short scripts first and pick one. Voices spend credits, so settle the words before making it.
- Record the screen slowly with big cursor moves; the punch-ins read much better on a calm recording.
New to prompting video? How to write a good prompt for Cut Express covers the story, the hook, captions, sound, pace and how to iterate with versions.
The real render behind this prompt
This prompt was written from one we used to make a video for this site. Here is that video and the exact prompt that made it.
The exact prompt
Make a 20-second 9:16 narrated video that tours the cut.express landing page, using the screen recording I uploaded. Inspect it first (contact sheets) and pick the moments where the hero, "How editing videos with Claude works", the feature cards and the pricing are on screen. Use fit: blur so the landscape recording sits in the vertical frame, with slow punch-ins on the parts the voice talks about.
Voice-over (make it with audio, a calm confident male voice, English), this exact script:
"You have the footage. Editing is the bottleneck. With Cut Express, you just tell Claude what you want. It reads your clips, picks every cut, caption and sound, and renders the MP4. Reels, Shorts, YouTube. Seven days free at cut dot express."
Over the recording, add a chat overlay as a motion graphic (animation overlay, transparent, box in the top half, x 0.06, w 0.88): a dark chat card (#131310, 1 px border #3A372C, radius 20) where a user bubble types "cut my 3 clips into a 30s reel. hook first, bold captions, warm look", then a reply line from Claude appears ("inspecting 3 clips... writing the timeline... rendering v1") with a small yellow #FFD23F progress bar. Keep it on screen from about 3 s to 12 s, with a slide_up entrance and fade_out exit.
Captions: source auto from the voice track, preset bold, highlight color #FFD23F, bottom third, 2 lines max, so they never cover the chat card.
Music: a soft beat under the voice if I uploaded one, ducked. Add a whoosh on each section change (built-in sfx).
End card (last 2.5 s, motion scene, background #0B0B0A): the cut.express wordmark and "Start your 7-day free trial".
Show me the timeline before you render.