POV: a 15-second first-person scene with the text doing the storytelling
The POV format in which the camera is the viewer's eyes and one line of on-screen text sets the situation; the prompt writes the text, the single action and the beat where it turns.
Write a POV script for {topic}. The format works because the viewer is cast as the character in the first frame and has to stay to find out what happens to them. No narration; the on-screen text and the picture carry everything. Structure (15 s, 9:16): - 0-2 s PREMISE: on-screen text, max 12 words, starting with 'POV:' and naming a situation the target viewer has been in. Camera already in the scene, first person, hands visible if relevant. - 2-9 s THE ACTION: one continuous first-person action, no cut if possible. No text. - 9-12 s THE TURN: the thing that changes (the result, the surprise, the small disaster). On-screen text, max 6 words. - 12-15 s BUTTON: a reaction beat held in silence, or one line of text that is a question. No CTA in this format; the product is in the scene, the comments ask about it, the answer is a comment. Word budget: 0 spoken words. On-screen text total under 25 words. Sound: ambient room sound or one trending-style sound placeholder marked SOUND. Write 3 premise variants (one relatable, one aspirational, one slightly absurd) and pick one. Return: the chosen premise, a 4-row table (time, what the camera sees, on-screen text, sound), and a 1-line note on how the phone is held. Topic: {topic} Target viewer: {who they are, the situation they recognise}
In Claude Code, paste the prompt as a message.
Example
Topic: a 10-minute-a-day language-learning habit. Target viewer: someone who has restarted the same app four times.
Premise variants:
- POV: it’s day 31 and you still open the app before Instagram (relatable)
- POV: you order in Portuguese and the waiter does not switch to English (aspirational)
- POV: your streak reminder goes off during the wedding vows (absurd)
Chosen: 2.
| Time | Camera sees | On-screen text | Sound |
|---|---|---|---|
| 0-2 s | A café table, menu in hand, waiter approaching | POV: you order in Portuguese and the waiter doesn’t switch to English | café ambient |
| 2-9 s | Hands put down the menu, point at an item, a nod from the waiter | (none) | café ambient, your voice low and off-mic |
| 9-12 s | Waiter walks away, the phone on the table lights up with a streak notification: 31 | he understood. day 31. | ambient |
| 12-15 s | Hands pick up the coffee, held | 10 minutes a day. that’s all it was. | ambient |
Phone held at chest height with both hands, slightly tilted down so the table and hands are in frame.
Tips
- The premise is the hook. Spend most of the time on it; the action is just proof.
- One long take beats three cuts here. A cut breaks the 'you are here' illusion.
| works with | Claude Code: Runs as a skill or a prompt in the terminal; can read your files and write the result. · Claude: Paste into claude.ai or the Claude app. · ChatGPT: Paste into ChatGPT; works with any GPT model. · Any LLM: Model-agnostic: Gemini, Mistral, a local model. |
| cost | Free. No API, no key. |
| status | Unverified: written from the technique, not yet run end to end by us. |
| license | CC BY 4.0 for the text here; skills are MIT. |
| updated | 2026-11-01 |
| for agents | pov-format-script.md |
Related
Reaction / duet style: respond to a clip in 30 seconds without stealing it
The split-screen or picture-in-picture response format, written so the reaction adds a fact or a demonstration instead of a face. The prompt plans the pause points and what you say at each.
Silent text-on-screen story: 25 seconds, one line per shot, paced for reading
The no-voice format that works with sound off because the text is the narration; the prompt paces each line to its shot at 2 words a second and puts the turn on the second-last card.
Before/after: the 20-second reveal with the 'after' shown first
The before/after format with the twist that keeps it from being skipped: open on the finished result, then show how bad it was, then the change. The prompt returns a timed script and text plan.
Day in the life: a 45-second cut from real footage with one through-line
A recipe for the vlog-style day that keeps working because the viewer waits for one promised payoff; the prompt turns your clip list into a timed script with text overlays.
Countdown listicle: 5 items in 45 seconds, best one last, numbers on screen
The ranked list format, with the rule that makes it hold: each item gets shorter as the count drops and number one gets the longest shot. The prompt returns script, text plan and the caption list.
'Stitch this' opinion: 5 seconds of their claim, 25 seconds of your answer
The stitch format in which a short borrowed clip sets up a claim and you answer with one argument and one demonstration. The prompt picks the 5 seconds to keep and bans the straw man.