format-hooks
Skill0 starsChanged 2 months agoMITView on GitHub
Hook frameworks, the 10-variation generation method, platform retention windows, and hook scoring. Use when mkt-content needs a hook menu for a reel, LinkedIn post, or carousel cover before building the full asset.
In this repository20 other files it gives its agents.
- CLAUDE.mdCLAUDE.md
- global/CLAUDE.mdCLAUDE.md
- copywriting-frameworksSkill
- design-mdSkill
- format-carouselSkill
- format-linkedinSkill
- format-reelsSkill
- funnel-designSkill
- humanizationSkill
- launch-playbooksSkill
- offer-designSkill
- pipeline-brandSkill
- pipeline-campaignSkill
- pipeline-contentSkill
- pipeline-copySkill
- pipeline-landing-pageSkill
- pipeline-luxoSkill
- pipeline-refino-uiSkill
- pipeline-sistemaSkill
- scroll-worldSkill
The file
We split this skill into the rules it gives an agent
2 rules. About 0 are common, 1 is distinctive, and 1 is very specific to this project.
≈0
Common · estimate
Many files like this have a rule of this kind.
≈1
Distinctive · estimate
This author's own choices. Worth a look.
1
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Distinctive: line 39
Check angle coverage: at least 2 of the 4 persuasion angles (more-good / less-bad / status / ease) represented across the set.
- Very specific: line 48
No fake urgency/scarcity per the offer-design ethics section.
≈ “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.
Files
Format: Hooks
Hooks are the first 2-8 seconds of video, or the first 1-2 lines of text/carousel cover, that decide whether the audience keeps going or moves on. This skill governs hook production for any format, consumed by mkt-content and referenced by mkt-copywriter for ad hooks.
Platform Retention Windows (design to the shortest one that applies)
| Platform / surface | Window |
| Short-form vertical video | 2-3 seconds |
| Long-form vertical video | up to ~8 seconds — can breathe slightly, still front-load the point |
| Feed text post (before truncation) | ~2 lines / roughly the first 140 characters |
| In-feed video | ~3 seconds — same discipline as short-form vertical |
| Carousel cover slide | ~1 second visual + one line of text |
These exact numbers shift as platforms change their UI and autoplay behavior — verify current truncation/retention specifics if it's been a while since last checked. The discipline itself (front-load everything) does not change even when the second count does.
10 Hook Types
- Contrarian claim — states the opposite of common belief.
- Startling statistic — a number that reframes the stakes.
- Personal/specific story — a concrete incident, never a generality.
- Pattern interrupt — a visual or verbal break from the expected opener.
- Controversial question — one the audience can't help but answer internally.
- Specific outcome promise — names the result and the timeframe.
- Wrong-way reveal — "most people do X backwards, here's why."
- Before/after contrast — two concrete states, no vague middle.
- Authority drop — a credible pattern observed across many cases.
- Curiosity gap — names that something exists without revealing it yet.
10-Variation Generation Method
- Confirm the brief: topic, objective, audience, platform, and angle material from
copywriting-frameworks(at least 2 persuasion angles). - Generate one hook per type above — 10 minimum, forcing type variety rather than 10 rewrites of the same idea.
- Check angle coverage: at least 2 of the 4 persuasion angles (more-good / less-bad / status / ease) represented across the set.
- Discard-and-regenerate rule: if 3 or more hooks in the set feel like minor rewrites of each other, replace them before delivering — the set must offer a real choice, not the illusion of one.
- For each hook, note: type, angle, target platform, and a one-line reason it should work (or where it might miss).
Hook Scoring (quick filter before handoff)
- Lands inside the platform's retention window when read/watched at natural pace? (pass/fail)
- Passes the specificity test — could this line only be about this topic, or could it be about anything? (concrete beats generic)
- Zero banned words from the
humanizationskill. - No fake urgency/scarcity per the
offer-designethics section. - Readable or hearable in one breath (video) or one glance (text/cover).
Output Format
# Hooks — <Topic>
Objective: <> | Audience: <> | Platform priority: <>
Angle material:
- More-good: <> | Less-bad: <> | Status: <> | Ease: <>
1. <text> — Type: <> — Angle: <> — Platform: <> — Why it works: <>
2. ...
[through 10+]
Recommendation: <which hook for which context, and why>Handoff
The selected hook goes verbatim into format-reels (as the script's opening), format-linkedin (as the post/carousel cover opening), or directly to mkt-copywriter for ad copy. Hooks never ship standalone — they are always the opening beat of a larger downstream deliverable.
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
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.

