YouTube

Verified against ChatGPT · 2026-08-12

Turn one long-form video transcript into a platform-specific repurposing plan, not just clipped copies

Analyzes a transcript to find the moments actually worth extracting, then maps each one to the platform and format it fits best, instead of chopping the same three clips into every channel.

ChatGPT (GPT-5.1)3 fillable variables

The prompt

Ready to copy — highlighted parts are example details you can swap.

You are building a repurposing plan from one piece of long-form content, not just picking a few 'best clips' and posting the same clip everywhere. Different platforms reward different things, and the moment that works as a YouTube Short might fail on LinkedIn and vice versa — I need you to reason about that fit, not just extract highlights.

SOURCE CONTENT
42-minute podcast interview transcript with a nutritionist debunking three common protein-intake myths, including a specific study citation and a personal anecdote about a client.

PLATFORMS I ACTUALLY POST TO
YouTube Shorts, Instagram Reels, LinkedIn (text posts), and the newsletter.

WHAT ALREADY WORKS FOR ME ON EACH
Shorts do best with a contrarian claim in the first line; LinkedIn posts do better as a numbered myth-busting list than as a story.

STEP 1 - FIND THE MOMENTS
Read the source and identify 5-8 distinct moments worth extracting: a strong claim, a demonstration, a specific number or result, a disagreement or correction of a common belief, a personal story beat. For each, note the timestamp or location in the source and, in one line, why it stands on its own without the surrounding context.

STEP 2 - MAP TO PLATFORM, NOT THE OTHER WAY AROUND
For each moment, decide which platform (if any) it actually fits, based on the platform's real constraints — a moment that needs 45 seconds of buildup to land doesn't belong on a platform where the first two seconds decide everything; a moment that's a specific number or result with no visual dependency can become a text-first post; a moment that only makes sense with the original speaker's tone or face doesn't translate to a caption-only format. Do not force every moment onto every platform — some moments should map to only one, some to none, and say so plainly rather than padding the plan.

STEP 3 - ADAPT, DON'T JUST TRIM
For each moment-to-platform pairing, note what actually needs to change beyond the length — the hook line, whether text overlay replaces spoken context, whether the CTA needs to differ.

WHAT NOT TO DO
Do not recommend posting the identical clip with only the aspect ratio changed across all platforms — that's not a repurposing plan, it's just resizing. Do not manufacture a platform fit for a moment that doesn't have one just to hit a quota.

OUTPUT FORMAT
A table: Moment | Source location | Best-fit platform (or 'none') | What needs to be adapted | One-line hook rewrite for that platform. End with a short note on which platform got the fewest strong moments and whether that's a source-content gap or a platform-fit issue.

Customize

Optional — swap in your own details for the highlighted parts above.

Why this works

The default failure mode in repurposing — and the one GPT-5.1 falls into if just asked to 'suggest clips for social media' — is treating platforms as interchangeable output formats that differ only in aspect ratio and length, when in fact each platform's own recommendation system rewards a different kind of moment: a short-form vertical feed rewards a moment that lands its point in the first two seconds with no buildup dependency, while a text-based professional platform rewards a moment that can be argued or listed without needing tone or visual context at all. Forcing the model to find moments first and only then reason about platform fit, rather than starting from 'here's five clips for Instagram,' prevents it from bending a moment to fit a platform it doesn't actually suit, which is the exact 'resize and repost' pattern that produces content that technically exists everywhere but underperforms everywhere because it was optimized for none of them. Explicitly permitting 'none' as a valid platform fit matters because a model asked to fill a repurposing plan will, by default, try to make every extracted moment useful somewhere to seem maximally helpful, which produces forced, weak pairings; giving explicit permission to say a moment doesn't repurpose well removes the pressure to pad the output and keeps the recommendations honest. The final gap-diagnosis question — is a platform's weak showing a source problem or a fit problem — matters practically because the fix is different in each case: a source gap means the next long-form piece needs to be recorded with that platform's needs in mind, while a fit problem means stop trying to force that platform from this kind of source content at all.

What you get back

Moment: the specific 1.6g/kg protein figure and its study citation. Source: 8:40. Best-fit: LinkedIn (text-first). What needs adapting: strip the anecdote framing, open with the number as a claim, cite the study inline rather than verbally. Hook rewrite: 'Most people are eating 40% more protein than they need. Here's the study.' Weakest platform: the newsletter got only one strong moment — this source leans conversational and story-driven rather than reference-heavy, which is a source-content gap, not a fit issue with the newsletter format itself.

Verified against

ChatGPT GPT-5.1 · 2026-08-12

Changelog

  • 2026-08-12 Initial publish, verified against ChatGPT GPT-5.1.

Need this built into your business?

If a prompt isn't enough — what Scult builds, built and maintained for you — that's Scult's day job.

EXPLORE WHAT SCULT BUILDS
All YouTube prompts

Check your AI visibility

One URL in, a 0–100 score and the exact fixes out.

RUN THE CHECK

Browse all the tools

15 tools across six categories
13 of them never send your data anywhere

Free · No signup · No trial clock

SEE THE DIRECTORY