agentleFS
scrollmarkscrollmark/social-skills/skills/hook-anatomy/SKILL.md

hook-anatomy

Skill19 starsChanged 29 days agoMITView on GitHub

Use when analyzing, evaluating, or generating hooks for short-form video or text posts.

The file

We split this skill into the rules it gives an agent

1 rule. About 0 are common, 1 is distinctive, and 0 are 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.

  • 0

    Very specific

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

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    Does it open a loop?: If the viewer can feel satisfied after the first 3 seconds, there's no hook.

    line 30

≈ “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

Hook Anatomy

A hook is the first 1-3 seconds of video (or first line of text) that creates a reason to keep consuming. It's not a summary. Not an introduction. A hook opens a loop the viewer needs closed.

Hook Patterns

These are structural patterns, not templates to copy verbatim.

Curiosity gap — implies knowledge the viewer doesn't have. "I found out why..." / "Nobody talks about this part." Works because humans are wired to close information gaps.

Pattern interrupt — visual or verbal surprise that breaks scroll autopilot. An unexpected image, a jarring first word, a scene that doesn't immediately make sense. The viewer stops to figure out what they're looking at.

Identity call-out — "If you're a [specific identity]..." Instant relevance filter. People who identify will stay; others scroll. High retention rate because the audience self-selects.

Contrarian take — "Unpopular opinion..." / challenges something the audience believes. Creates tension that needs resolution. Works best when the take is genuinely surprising, not just inflammatory.

Transformation tease — before/after implied. "I tried X for 30 days." The viewer needs to see the result. The bigger the implied gap between before and after, the stronger the hook.

Story drop — starts in the middle of action. "So there I was, standing in my boss's office, holding a pineapple." Context is missing and the viewer needs it.

Evaluating Hook Strength

Three questions:

  1. Does it open a loop? If the viewer can feel satisfied after the first 3 seconds, there's no hook.
  2. Could you scroll past? Be honest. If nothing about the first moment demands attention, it's weak.
  3. Is the payoff implied? The viewer should have a sense of what they'll get by staying, even if the specifics are unclear.

Platform Differences

Hooks work differently by platform. Load the relevant references/platforms/{platform}.md from this skill's own directory.

Key distinction: TikTok hooks are visual + text overlay (the first frame and on-screen text do the work). Instagram Reels hooks are often spoken — the first words matter most. YouTube hooks happen before the video — the thumbnail and title ARE the hook. First 30 seconds then justify the click.

Anti-patterns

  • "Hey guys, welcome back" — no open loop, no reason to stay. The viewer doesn't care about your greeting.
  • Clickbait without payoff — opens a loop but never closes it, or the answer is disappointing. Erodes trust and kills long-term retention.
  • Hook-content disconnect — the hook promises one thing, the content delivers another. Viewers learn not to trust you.
  • Slow hooks — if the hook takes more than 3 seconds to land on short-form video, it's too slow. Every fraction of a second matters in the first moments.

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.