Index of every page on this site: llms.txt
All skills
sergebulaev avatar

/tt-hook-scripter

@ba2f700
Run it yourself

Script the first 1-3 second TikTok hook: the spoken line, the on-screen text, and the opening visual firing at once so the viewer cannot swipe. Picks a 2026 hook formula (cold-open result, pattern interrupt, number reveal, open-loop question, relatable call-out, story) by goal of completion, saves, comments, or shares. The single biggest retention lever on TikTok. Use to script a hook from a video idea. Not for the caption (use tt-caption-writer) or scrubbing a script (use tt-humanizer).

Use this Skill: https://skilld.dev/gh/sergebulaev/tiktok-skills/tt-hook-scripter

Nothing lands on disk. Nothing to clean up.

Fork this Skill

Edit a local copy. It keeps the author and licence.

referenceshook-anatomy.md

≈821 tokens on demand. Your agent reads this file only when SKILL.md points to it.

Hook Anatomy: the three-layer open and the timing budget

The TikTok hook is not one thing. It is three things firing inside the first 1 to 3 seconds: the spoken line, the on-screen text, and the opening visual. Each does a different job, and they have to agree on one promise.

The three layers

1. The spoken line (sound-on promise)

What you say to camera before the viewer can swipe. Job: land the promise for the ~70% who have sound on. Rules:

  • One breath. If it needs two, it is too long for the open.
  • A contraction and a fragment beat a full grammatical sentence.
  • Say the number out loud ("forty-seven minutes"), do not just flash it.
  • Cut the throat-clear. The first word is the payoff or the tension.

2. The on-screen text (muted-first promise)

The text card in the first second. Job: carry the promise for the viewers who start muted, which is a large share of every video's first impressions. Rules:

  • 3 to 7 words. It gets read at scroll speed or not at all.
  • Different words from the spoken line. Saying the same thing twice wastes the second; two angles on one promise widens who it catches.
  • At most one em dash, and a card rarely needs one; minimal punctuation. A line break beats a comma.

3. The opening visual (frame one)

What is on screen in the very first frame. Job: stop the thumb with motion, result, or an out-of-place object. Rules:

  • Show the result (T1), the tension (T8), or an interrupt (T2) immediately.
  • No logo, no greeting card, no slow zoom-in. Dead air in frame one is the top completion killer.
  • Movement helps. A static talking head with no text and no motion is the weakest possible open.

The timing budget

Beat Time What happens
Frame one 0.0s the visual hits (result / tension / interrupt)
Spoken line 0.0-1.5s the sound-on promise lands
On-screen text 0.0-2.0s the muted promise is readable
The turn by 3.0s a reason to keep watching is established

If a reason to stay is not on screen by second three, the video is already losing the viewers it will never get back.

The loop-close (why the last frame matters as much as the first)

A rewatch counts almost like a fresh view, and a clean loop is the cheapest way to push average watch time past 100%. Design the ending so the first frame lands differently the second time:

  • End on the line that recontextualizes the open ("..and that is why the first clip looked wrong").
  • Cut the last frame so it butts cleanly against the first (a visual match cut).
  • Do not end on "thanks for watching" or a logo. That kills the loop.

Worked example

Idea: a productivity creator showing a note-taking system.

  • Formula: T1 Cold-Open Result (goal: completion)
  • Frame one (visual): the finished, color-coded note board already on screen
  • Spoken line: "this replaced every to-do app i tried in 2025."
  • On-screen text: "one note, zero apps"
  • The turn (by 3s): "here is the whole system in 4 steps."
  • Loop-close: end on "..and step one is the part everyone skips", which makes the viewer rewatch to catch step one.

Note the spoken line and the on-screen text say different things ("replaced every app" vs "zero apps"), both pointing at one promise.

Source: SKILL.md on GitHub

No rule matched.

skilld matched fixed text patterns in SKILL.md and file names. Patterns miss obfuscated code.

skilld run checks every file with the same patterns. It asks for approval before it loads a Skill with a behavior marked Needs approval.

No alerts1mo3 checks · Risk SAFE
  • Gen Agent Trust Hub1mo

    The skill is a specialized scriptwriter for TikTok video hooks. It uses established formulas to generate spoken lines, on-screen text, and visual descriptions based on user-provided video ideas. No security risks or malicious patterns were identified.

  • Socket1mo

    No alerts

  • Snyk1mo

    Risk: LOW · No issues

Signed by skilld at ba2f700. This ties the file your Agent reads to that commit on GitHub. It does not review the instructions.

Last checked against GitHub yesterday.

Activeupdated 2 months ago

README badge

README badge for sergebulaev/tiktok-skills/tt-hook-scripter