Stop the Scroll

Version 0.2.0. Free. Works with major AI agents. Source: github.com/diaryofamademan/skills

What it does

Most content is built as one piece, middle first. This skill builds it as three: a hook, a body, and a call to action, each held to its own standard, because the ends are where pieces die.

The frame is AIDA, the oldest structure in advertising: attention, interest, desire, action. The skill hardens its two ends with thirteen principles distilled from one real launch session.

The hook talks to the reader, not at them: a direct question about their own problem, framed so anyone can feel it. It has to stop a scroll among a hundred competing pieces, and it has to make sense screenshotted alone.

The middle stays universal and about impact: simple language, every number carrying its why, things named as what they are. Detail the reader does not need is cut, however interesting it is to the author.

A statistic chosen to carry a point follows a further rule: stats multiply or reduce, increasing money, increasing efficiency, reducing waste, or increasing speed. It has to pass three tests: one visible number, no reader math, and a change shown rather than a state.

Six more rules were earned in the same launch's later rounds, when its results post went through six more versions. A result is reported from the source itself, never from a summary of it, and a formula is never spelled out as a sentence: shown as math, three numbers with named rates and round percentages, it reads at a glance.

Each stage of a flow keeps one everyday word at every mention, so the reader can follow it across every post. Where a system leaks, the piece opens with what went in and what came out, and a claim to see what others miss is shown, not stated: smart actors missing the whole picture, each locally right and globally blind.

The call to action gets the same hardening: a jab with a point of view, not a generic ask. It stands alone as its own post, idea, name, and link, and the link is eased into, never dropped bare.

Two tests decide whether a piece ships: the sea-of-100 test and the stand-alone test, run on each end. A line that fails either test is rebuilt, not defended.

The skill ships with its proof: its worked example is the launch of Fix the Bottleneck, one piece of copy run through fourteen versions, each version killed by a principle on this list. See the skill it launched: Fix the Bottleneck

Use it when:

  • You are launching something in a feed: a product, a skill, a feature, an idea, as a post or a thread on X or LinkedIn.
  • You are writing a video script or an outline, where the first line decides whether anyone stays.
  • You are writing website or landing copy, or any announcement competing against a hundred other pieces posted the same hour.

What is inside:

  • The five step build procedure and the twelve guardrails: build the ends first, run both tests on each end before the piece ships, no bare numbers, name things as what they are, and keep the author's direction out of the copy.
  • The thirteen principles in full: for each one the rule, why it exists, the failure it prevents, and its quick test, plus the construction patterns for hooks and calls to action and how to run the two tests.
  • The worked case study: the fourteen-version launch session the principles were distilled from, with the final copy, and the output contract every build run and review run returns.

Get the skill

Put your email in and Stop the Scroll arrives right away: the install link and a one page quick start. You also join Running on Agents, my free newsletter, and get every update the day it drops.

Your email is used to send the skill you asked for and the free newsletter. The list runs on Beehiiv.

You can unsubscribe in one click in any email. Your address is not sold.

If you apply for the audit, that information is used only to review your application, unless you also tick the newsletter box on that form.