The prompt
Score 10 of 10, Excellent
Use it when You have a written article that does well and want a script for a YouTube video on the same topic that sounds spoken, not read aloud.
- {{channel}}
- {{audience}}
- {{length_minutes}}
- {{video_format}}
- {{visuals}}
- {{call_to_action}}
- {{article}}
Full prompt and why it works
Prompt
Turn the article below into a script for a YouTube video on {{channel}}, for {{audience}}. Viewers decide in the first seconds whether to stay, and they cannot skim back the way readers can, so the script has to be rebuilt for listening rather than read aloud from the page. Target length: {{length_minutes}} minutes How we film (talking to camera, voice over a screen recording, a demo at a desk, or slides with a voice over): {{video_format}} What we can show on screen: {{visuals}} The one thing viewers should do at the end: {{call_to_action}} The article: """ {{article}} """ Write: 1. A cold open of 15 to 30 seconds that shows or states what the viewer will get. Open with the article's most useful point, not its introduction. 2. The script in sections. For each: a working title, the spoken words, and on-screen notes in square brackets, such as [close-up of the form]. Write it all word for word, except the middle of a section, which may be bullet beats if that sounds natural. 3. A timing line per section, at about 150 spoken words per minute, adding up to the target length. 4. What you cut from the article, and why. How to rebuild it for video: - Spoken sentences are shorter than written ones. Use contractions, and say numbers the way people say them. - Replace headings with spoken signposts ("Here's the second thing...") and repeat the main point once near the end, since viewers cannot scroll back. - Turn tables and long lists into something we can show, or the few items that matter. - Where the article cites a source, name it in the script and list the link for the description. A good script should sound like one person explaining to one viewer, and someone who only watches the video learns what the article teaches. For example, a transition in the right form: "That fixes the setup. The bigger problem is what happens on day two." That example shows the form only. Keep every fact, number, quote and claim as the article states it, and do not add new statistics, examples, quotes or promises. If the article does not fit the target length, tell me what you would cut or move to a second video instead of rushing it. If a section needs a visual we cannot make, mark it [NEED VISUAL: what].
Variables
- {{channel}}
- Your channel and what it covers, for example "a channel for small landlords".
- {{audience}}
- Who watches and what they already know, for example "new freelancers who have never sent an invoice".
- {{length_minutes}}
- Target length in minutes, for example 8.
- {{video_format}}
- How you film, for example "talking to camera with screen recordings".
- {{visuals}}
- What you can show: product, screen, props, your own photos or charts.
- {{call_to_action}}
- The one action at the end, for example "download the checklist linked below".
- {{article}}
- The full article, pasted as published.
Why it works
- What to make. The job is a rebuild for listening, not a read-aloud: signposts replace headings and tables become something to show.
- What to avoid. Every fact and claim stays as the article states it, so the video cannot say more than the article you already checked.
- If unsure. An article too long for the target is cut or split openly, and a section needing a visual you cannot make is marked rather than invented.