X (Twitter)

Verified against ChatGPT · 2026-08-13

Rewrite a flat opening line into ten hooks that survive the timeline's first half-second

Takes a post you've already drafted and generates ten alternative first lines built around distinct hook mechanisms, each tagged with why it earns attention, instead of ten variations of the same curiosity-gap phrasing.

ChatGPT (GPT-5.1)2 fillable variables

The prompt

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

You are a specialist in X/Twitter hook-writing. I already have a post drafted — the argument and content are done. Your only job is the first line: generate ten alternative opening lines, each using a genuinely different hook mechanism, not ten reshuffles of the same one.

DRAFTED POST (content is final, only the opener is up for revision)
...we tested 4 pricing pages against the same traffic. The one with the fewest features listed converted 22% higher than the one with the most.

WHO THIS NEEDS TO STOP-SCROLL
SaaS founders currently deciding between a feature-heavy or minimal pricing page redesign.

HOOK MECHANISMS TO DRAW FROM (use at least 6 of these across your 10, don't default to one):
- A specific, surprising number pulled from the post's own content
- A contradicted assumption stated as fact, then immediately undercut
- A direct address to a specific identity in the target reader
- A confession of a mistake or wrong belief you used to hold
- A blunt claim that sounds like an overstatement until the reader checks the content
- A question that only the target reader would stop to answer honestly

RULES
Every hook must be traceable to something actually in the drafted post — do not invent a number, result, or claim that isn't already there just to make an opener punchier; if the post doesn't contain a strong enough number or result to hook on, say so and tell me which hook mechanisms are viable given what's actually in the draft. After each hook, add a one-line tag naming which mechanism it uses and, in a few words, why it would stop that specific target reader rather than a general audience. Do not produce two hooks that are the same mechanism with different word choice — if you can't genuinely fill six distinct mechanisms from the list given the content, produce fewer, correctly labeled, rather than padding to ten with near-duplicates.

WHAT NOT TO DO
Do not write hooks that promise something the rest of the post doesn't deliver — a hook is only as good as whether the payoff actually lands, and an opener that oversells the content gets read as clickbait and hurts the account's credibility on the next post. Avoid "Twitter voice" clichés ("unpopular opinion:", "nobody talks about this but", "hot take") unless the post's content genuinely is a contrarian take that earns that framing.

OUTPUT FORMAT
A numbered list of up to 10 hooks, each followed by its mechanism tag and one-line reasoning. End with your top 2 picks and a one-sentence reason each is the strongest match for the target reader specifically.

Customize

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

Why this works

Requiring at least six distinct hook mechanisms rather than letting the model free-associate ten openers matters because GPT-5.1's unconstrained default, when asked for hook variations, tends to converge on one or two high-frequency patterns from its training distribution (contrarian-claim framing and curiosity-gap framing especially) and produce nine reshuffles of those rather than genuinely different attention mechanisms — naming the mechanisms explicitly and requiring a minimum spread forces actual variety instead of superficial wording changes around the same idea. Tying every hook back to content that's actually in the drafted post closes the most damaging failure mode in hook-writing: a hook so much stronger than the payoff that the reader feels baited, which on X specifically tends to produce a reply calling out the mismatch rather than silent scrolling past, since the platform's reply-first culture actively rewards catching an overpromise. The mechanism tag with reasoning is not decoration — it forces the model to justify why a given hook fits the stated target reader rather than a generic audience, which surfaces the difference between a hook that's clever in the abstract and one that would actually make the specific person named in target_reader stop on their specific timeline, where a founder deciding on pricing page design reacts to a different signal than a general marketing audience would.

What you get back

1. "We removed 60% of our pricing page's feature list. Conversion went up 22%." — Mechanism: surprising number from the post's own content. Stops founders currently weighing feature-density decisions because the number contradicts the default instinct to list more, not less. 2. "I used to think a pricing page's job was to prove we had every feature. It isn't." — Mechanism: confession of a former wrong belief. Lands with founders who currently hold that same belief right now. Top picks: #1 for its directness with a founder audience that responds to numbers over narrative; #2 for founders earlier in their own pricing-page debate who haven't seen a result yet.

Verified against

ChatGPT GPT-5.1 · 2026-08-13

Changelog

  • 2026-08-13 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 X (Twitter) 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