agentleFS
sergebulaevsergebulaev/tiktok-skills/skills/tt-trend-mapper/SKILL.md

tt-trend-mapper

Skill103 starsChanged 5 months agoMITView on GitHub

Map a trending TikTok sound or format to your niche so the ride is authentic, not cringe. Runs a fit check (does it match your niche, is it still early, what is the native joke), then scripts the twist that makes the trend yours: the recognizable beat plus a niche-specific turn. Covers sound timing, the trend lifecycle, and when to skip a trend entirely. Use to ride a trend without looking late or forced. Not for a from-scratch hook (use tt-hook-scripter) or a posting plan (use tt-content-planner).

The file

We split this skill into the rules it gives an agent

6 rules. About 0 are common, 5 are distinctive, and 1 is very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈5

    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
  1. Very specific:

    Before tt-hook-scripter when the hook is a trend-ride (formula T10)

    line 19
  2. Distinctive:

    A skip recommendation: when the trend does not fit (saying no is a feature)

    line 28
  3. Distinctive:

    If a trend fails the fit check, recommend skipping it.

    line 45
  4. Distinctive:

    Keep the native structure: If the trend has a cut on the beat drop, cut on the beat drop.

    line 69
  5. Distinctive:

    Read the lifecycle stage: Ask the user how saturated they have seen it.

    line 82
  6. Distinctive:

    Never fabricate a trend's popularity.

    line 101

≈ “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

TikTok Trend Mapper

Riding a trend is a distribution lever: a trending sound carries built-in reach, and a familiar format lowers the barrier to watch. But a trend copied with no twist is forgettable, and a trend forced onto a clashing niche reads as cringe and costs trust. This skill decides whether to ride, and if so, scripts the twist that makes the trend yours.

When to use

  • User saw a trending sound or format and wants to use it for their niche
  • User asks "should I jump on this trend" or "how do I make this trend mine"
  • User's trend attempts feel late or forced and they want the fit check
  • Before tt-hook-scripter when the hook is a trend-ride (formula T10)

What this skill produces

  • A fit verdict: ride it, bend it hard, or skip it (with the reason)
  • The trend teardown: the native joke / structure / beat, and why it works
  • The twist: the recognizable beat plus the niche-specific turn, scripted as a hook (spoken line + on-screen text + visual)
  • A timing read: is the trend still early, peaking, or stale
  • A skip recommendation when the trend does not fit (saying no is a feature)

The fit check (run this first)

Four questions. A trend has to clear all four to be worth riding.

  1. Does it match your niche? Can your audience tell why you, specifically, did this trend? If the only connection is "it was trending", skip it.
  2. Is it still early? A sound in its first 1 to 3 days of climbing carries the most lift. If it is already on every third video, you read as late. Check how saturated the sound and format already are.
  3. Is there a native structure to honor? Most trends have an expected beat (a setup line, a specific cut, a punchline placement). Riding it means hitting that beat, then twisting it, not ignoring it.
  4. Can you add a real twist? The twist is the whole point. If you cannot bend the trend to say something only your niche would say, the ride is filler.

If a trend fails the fit check, recommend skipping it. A skipped trend costs nothing. A cringe ride costs trust.

The trend lifecycle

StageWhat you seeMove
Emerginga sound or format on a few breakout videos, climbing fastride now, this is the window
Peakingit is everywhere, big accounts are on itride only with a strong twist, expect less lift
Saturatedevery third video uses it, the joke is wornskip, or subvert it (do the opposite)
Deadit stopped climbing days agoskip; using it now reads as late

The reach boost lives in the emerging stage. By saturation the sound is just a sound.

Scripting the twist

Once a trend clears the fit check, build the ride as a hook (formula T10):

  1. Hit the recognizable beat first. The first second has to read as the trend, or you lose the built-in recognition.
  2. Twist into your niche by the turn. "{the trend's expected setup}, but for {your niche}" is the core move. The twist is the surprise that makes it shareable inside your audience ("they did the trend, but for us").
  3. Keep the native structure. If the trend has a cut on the beat drop, cut on the beat drop. Honoring the format is what signals you get it.
  4. Hand the spoken line to tt-humanizer so the script does not sound written. Hand the full hook to tt-caption-writer for the caption.

Steps

  1. Gather inputs. The trend (a sound link, a format description, or an example video URL), the user's niche, and what they usually post.
  2. Parse any URL. lib.url_parser.parse_tiktok_url(url) extracts the handle and video id from an example video.
  3. Run the fit check. Score all four questions. If it fails, recommend skip and explain why (saying no honestly builds more trust than a forced ride).
  4. Read the lifecycle stage. Ask the user how saturated they have seen it; place it on the emerging-to-dead scale.
  5. Tear down the native structure. Name the beat, the joke, the expected cut.
  6. Script the twist as a T10 hook (spoken line + on-screen text + visual).
  7. Humanizer pass on the spoken line (tt-humanizer V3: AI vocab by density, stacked triads, reveal bridges, sincerity openers; make it sayable, never insert a punch line).
  8. Hand off. tt-caption-writer for the caption and a sound-credit note; tt-content-planner if the trend ride is one slot in the week.

Hard rules

Global voice rules: see root SKILL.md Voice rules. Additional skill-specific rules:

  • A trend with no niche fit gets a skip recommendation. Do not force it.
  • The twist is mandatory. A trend copied straight is filler.
  • Ride early or do not ride. A stale trend reads as late.
  • Honor the native structure (the beat, the cut, the joke placement).
  • Never fabricate a trend's popularity. If you do not know the saturation, ask.

Anti-patterns (skill will refuse)

  • "Just do the trend because it is trending" with no twist and no fit.
  • Forcing a sound onto a niche it clashes with.
  • Riding a sound that is already saturated and calling it timely.
  • Ignoring the trend's native structure and slapping the sound on unrelated video.
  • A twist that punches down or mocks the original creator.

Resources

  • ../../references/algorithm-heuristics.md - sound selection, the trend window, original audio
  • ../../references/hook-formulas.md - T10 Trend-Ride With A Twist (three layers)
  • references/trend-fit-rubric.md - the scored fit check and lifecycle examples

Related skills

  • tt-hook-scripter - script the trend ride as a T10 hook
  • tt-caption-writer - caption and sound credit for the ride
  • tt-content-planner - slot trend rides into the week without over-relying on them

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.