agentleFS
PicsArtPicsArt/gen-ai-skills/skills/retention-hooks/SKILL.md

retention-hooks

Skill6 starsChanged 19 days agoMITView on GitHub

Keep a short video watched to the end with open loops - a hook that opens a question in the first seconds, an intensifier, a rehook at the middle and a payoff that closes them, whether the lines come from the user's footage or are written for a voice or the screen. Use when ordering the moments of a short cut from talking footage, or writing the lines a short puts on screen or in a voice.

The file

We split this skill into the rules it gives an agent

9 rules. About 0 are common, 9 are distinctive, and 0 are very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈9

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 0

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    Never open a loop the video cannot close, and never close one before the next is open.

    line 14
  2. Distinctive:

    20–30 s: all three when the material has a second question.

    line 31
  3. Distinctive:

    Over 30 s: all three. Without a rehook the middle loses the viewer.

    line 33
  4. Distinctive:

    When the lines come from what people said (a reel, a short, highlights), the loops are chosen, not written.

    line 37
  5. Distinctive:

    Place it at the middle even when the speaker said it earlier, as long as it stands on its own: a cold reader of the heard script says …

    line 45
  6. Distinctive:

    When the lines are written, for the screen or for a voice:

    line 54
  7. Distinctive:

    If one blank pulls a different way, change it.

    line 82
  8. Distinctive:

    The ending loops back: When the video ends without a call to action, its last frame leads into the first, for instance by ending on the …

    line 111
  9. Distinctive:

    Before the plan goes to the user, list each loop in one line: what it opens, the second it opens, the second it closes.

    line 120

≈ “Common” is an estimate: at least 1 in 5 of the 350,495 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
SKILL.md

Story loops

Hook, body and payoff tell you where things go. They do not keep anyone watching. What keeps a viewer is an open loop: a question the video has raised and not yet answered. A short holds as long as one loop is open, and sags wherever none is.

A loop is two moments: the line that opens a question and the later one that closes it. Never open a loop the video cannot close, and never close one before the next is open. A serial episode's exit question follows drama-series-writing instead.

The three loops

LoopOpensWhereIt works when
Hookthe first question: a result, a contrast, a problem, a claim the viewer wants provedin the first 2 sthe viewer can say what they are waiting to find out
Intensifierthe same question, sharper: a stake, a number, a "and it gets stranger", the reason it is harder than it looksbefore 15 s, and inside the first quarter of a short under 60 sit builds on the hook's question instead of starting another
Rehooka new question, just as the first ones start to fadenear the middle, from 35% to 60% of the lengthit is new: a second problem, a twist, the part the viewer did not expect, never the hook said again

The payoff closes the hook's loop, at or near the end. The rehook's loop closes before or with it. The serial exit question above may remain open.

Scale them to the length:

  • Under 20 s: the hook and the intensifier. There is no middle to sag; the payoff comes at once.
  • 20–30 s: all three when the material has a second question; otherwise the hook and the intensifier, and say so.
  • Over 30 s: all three. Without a rehook the middle loses the viewer.

Choosing them from footage

When the lines come from what people said (a reel, a short, highlights), the loops are chosen, not written. From the pool of moments:

  • Hook: the sentence that raises the question the rest answers: a surprising result, a strong claim, a problem said plainly. A moment the watch rated hook is a candidate; it must still open a question, not just sound strong.
  • Intensifier: a sentence right after it that raises the stakes of the same question.
  • Rehook: a sentence that turns: "but", "except", "what nobody tells you", a second problem. Place it at the middle even when the speaker said it earlier, as long as it stands on its own: a cold reader of the heard script says whether it does, not you (the speech-aware-cutting skill).
  • Payoff: the sentence that answers the hook. Never use the payoff as the hook: it closes the loop before it opens. Lead with the question, end with the answer.

Clean sentence edges still come first. A loop never justifies a clip cut mid-sentence.

Writing them

When the lines are written, for the screen or for a voice:

  • Hook: show the result and hold back how it was made, so the rest of the video answers "how?".
  • Intensifier: the first line after the hook makes the question sharper, with a stake.
  • Rehook: the line nearest the middle opens what comes next instead of describing what was: the harder request, the step the viewer did not expect.
  • Payoff: the last result answers the hook. The call to action comes after it, never instead.

The writing's own rules still hold: no hype words, nothing the footage does not show.

The hook

The hook buys the next two seconds. A viewer stops for something they already feel, never for a product. Four rules, each one checkable on the hook's frames and words:

  1. Open on a moment or a problem the viewer already feels. A frustration they have had, a result in the middle of happening, a "before" they recognise. Never a product shot, or a character posed and lit, with a slogan over it: the product comes in as the answer, once the viewer has a reason to want one. Giving that reason is the hook's job. If a viewer could not say why they should care, it is not a hook yet.
  2. The words sound like a creator talking. "I" or "you", in plain words, the way you would tell a friend: "every shot in my film was a wide". Never a tagline: no two short lines that mirror each other ("Rough notes in. A pitch deck out."), no claim about the product, and no brand name in the first second. Read the line aloud, and rewrite it if it sounds like a commercial. Sometimes use no words at all. When the picture alone raises the question, let it.
  3. One promise. The picture, the words and the payoff promise the same thing. Before building, write it as one sentence: "The viewer sees ___, reads ___, and by the end gets ___." If one blank pulls a different way, change it. The payoff keeps the promise exactly; a hook that promises more than the video shows is bait.
  4. Native, not produced. It looks like something a person posted: the real screen, the real footage, framed the way a phone frames it. Save the polish for the result.

Choose the opener from what the material holds:

OpenerIt opens onFor example
The problema frustration the viewer has had, in their own words"my AI film had zero close-ups"
POVthe viewer's own situation, in the second person"POV: your storyboard is all wide shots"
In the middlethe result happening, before any setup, which lands within the next 3 secondsthe shot list rewriting itself
Before and afterthe two states, joined by a hard cutthe wide still, then the close-up
The surprisesomething the viewer did not expect to see herea man in his fifties vaulting a wall for his coffee

Ad tells

Any one of these makes a short read as an ad, and the viewer scrolls past it:

  • the first frame is a product, or a character posed and lit like a commercial, with words over it;
  • a slogan: two lines that mirror each other, a benefit claim, a rhyme;
  • a brand, a logo or an app name in the first second;
  • a feature named before the viewer has a reason to want it ("New:", "Introducing");
  • words that grade the product (amazing, effortless, game-changing), a "!", or a price;
  • a call to action before the payoff, or one that interrupts it;
  • a voice that announces from a script instead of telling;
  • studio polish on the opener: perfect light, a seamless backdrop, a staged lifestyle shot;
  • more to read in the first 2 s than one short line.

The ending loops back. When the video ends without a call to action, its last frame leads into the first, for instance by ending on the problem the hook opened on, now solved. A replay then reads as one continuous video. A serial exit follows drama-series-writing instead.

The quality-review skill's cold watch tests a hook on the built draft: its first two pieces (to 3 s) say whether strangers who know nothing of the video keep watching.

Check the loops

Before the plan goes to the user, list each loop in one line: what it opens, the second it opens, the second it closes. Then read it back:

  • Does a loop open in the first 2 s?
  • Is some loop open at every second until the payoff, or is there a stretch where the viewer has nothing left to wait for?
  • Does every loop close? A serial exit question may stay open.
  • Does the hook break none of the four rules above, and use none of the ad tells?

A missing loop is a note, not a block: when the material has no second question, build without a rehook and say so in one line.

Download now (ZIP)

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

No reviews yet

People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.

  • It worked
  • Partly
  • It didn't
What did you use it for? e.g. Reviewing pull requests in a Rails app

Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.

Your trail

Files you open will collect here, so you can find them again.

Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.

Now do this for your own.

Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.