Everyday Writing

Verified against ChatGPT · 2026-08-12

Turn a messy, real work anecdote into a social post that doesn't read like every other founder story

Takes your rough, unedited memory of something that actually happened at work and shapes it into a platform-ready story post with a real narrative arc — without sanding it down into the same three LinkedIn cliché openers everyone recognizes and scrolls past.

ChatGPT (GPT-5.1)5 fillable variables

The prompt

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

You are shaping a real thing that happened into a social media story post. I'm going to give you the anecdote roughly as I remember it — out of order, missing some details, probably longer in the wrong places and shorter in the important ones. Your job is to find the actual narrative inside it, not to summarize it or dress it up with generic hooks.

THE RAW ANECDOTE (as I remember it, unedited)
So we had this client who wanted to cancel like two days before their product launch, they were panicking about a bug we hadn't even confirmed was real yet, I remember it was a Friday night and I was the only one who saw the Slack message, turned out to be a caching issue on their end not ours, but I didn't know that yet when I called them

WHERE THIS IS GOING
LinkedIn text post, single column, no image — needs to work broken into short line-break paragraphs.

THE POINT IT NEEDS TO LAND ON
That staying reachable on a Friday night, even when it's not technically your job, is what actually saves client relationships — not the fix itself.

WHAT'S OFF LIMITS TO REVEAL
Don't name the client or the product, don't imply which coworker didn't see the Slack message first.

OPENING LINE CONSTRAINT
Open with a concrete detail or moment, not a question or a claim about what 'nobody' talks about.

HOW TO BUILD THE STORY
First, read through the raw anecdote and identify where the actual tension is — the moment something was uncertain, at risk, or about to go wrong, even if I buried it in the middle of my notes. A story post that opens with the setup and only gets to the tension in paragraph three has already lost the scroll; reorder the events around that moment, don't just retell them chronologically. Cut every detail that doesn't serve the specific takeaway I gave you, even if it's a detail I clearly liked — a story with three morals is a story with none, and "interesting but irrelevant" detail is the single biggest reason these posts run long and land soft. Write the opening line to the exact constraint I gave you, and if the most natural opening line you can find violates that constraint, tell me instead of quietly writing around it. Do not resolve the tension too early — hold the reader in the uncertain middle for at least two sentences longer than feels comfortable, since a story that gives away the outcome in the second line has nothing left pulling the reader down. End on the specific takeaway, stated as something the reader can actually use or recognize in their own situation, not as a moral wrapped in a hashtag.

WHAT NOT TO DO
Do not open with a rhetorical question, "nobody talks about," a bolded contrarian one-liner, or any other opener that's become a recognizable template on its own — if my raw anecdote already has a more specific, less generic first line buried in it, surface that one instead. Do not pad the piece with a listicle of "lessons learned" at the end; one takeaway, stated once, is stronger than three stated as bullets. Do not invent details, dialogue, or outcomes that weren't in what I gave you — if a beat is missing and the story genuinely needs it, ask me for it rather than filling it in with something plausible-sounding.

OUTPUT FORMAT
1. The finished post, formatted for the platform I named (line breaks, length, and structure appropriate to it).
2. One line naming which real moment you built the tension around, so I can confirm you picked the right one.
3. If the hook constraint conflicted with the strongest available opening line, say so and show me both options.

Customize

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

Why this works

GPT-5.1's default move on a request like 'turn this into a LinkedIn post' is to reach for the format it has seen most densely represented in training data for that exact request — the rhetorical-question hook, the chronological retell, the closing bullet list of 'lessons learned' — because those are the statistically safe, most-reinforced shape of a 'story post' as a genre, not because they're actually the strongest version of your specific story. Explicitly banning that named set of openers in the WHAT NOT TO DO section forces the model off its highest-probability default and back into the raw material you actually gave it, which is the only place a genuinely non-generic opening line can come from. The instruction to reorder around the tension rather than retell chronologically matters because models tend to preserve the sequence information was given in unless told otherwise — a chronological brain-dump fed in raw comes back as a chronological post unless the prompt explicitly authorizes restructuring, which is why the anecdote is deliberately supplied out of order here rather than pre-organized. Naming one single takeaway rather than letting the model infer 'the lessons' is what prevents the closing-bullet-list failure mode: an unconstrained model asked to wrap up a story will often hedge by offering two or three possible takeaways so it can't be wrong about which one you wanted, which is exactly the flattening effect that makes so many of these posts forgettable. The instruction not to invent missing beats matters mechanically too — a model filling narrative gaps with plausible-sounding invented dialogue or detail is the single fastest way a personal story stops being true, and asking the model to flag the gap instead keeps the authorship and factual accuracy with you rather than quietly delegating it.

What you get back

Friday, 9:47pm. A Slack message from a client I almost didn't see. Two days before their launch, they wanted out — convinced a bug we hadn't even confirmed was ours was about to blow up their weekend. [...] I picked up the phone not because it was my job, but because nobody else had seen the message yet. It turned out to be a caching issue on their end. That part didn't matter. What mattered was that someone answered. That's the part of client work nobody puts in the deck: the fix is rarely what saves the relationship. Being reachable when it counts is.

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 Everyday Writing 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