Token导航 LogoToken导航TokenDH.com
研究检索只读github未标认证来源可访问许可证需确认审计通过

creative-mechanics创意机制

Agent Skill

creative-mechanics 用于查找、检索和筛选相关信息,适合在 Codex、Claude、Cursor、Gemini CLI 中需要根据关键词、任务场景或来源线索快速定位候选结果时使用。可结合来源仓库、安装命令和原始 README 继续核验具体用法。安装前建议确认权限范围、维护状态,以及是否会触发联网、命令执行或文件读写。

总安装

3,856

周安装

156

GitHub Stars

88

下载量

1,211
CodexClaudeCursorGemini CLI

安装说明

本站只整理中文说明和来源信息,不托管安装包,也不代用户安装。

GitHub

来源数

2

许可证

unknown

最后核验

2026-05-01

来源状态

来源可访问

安装方式

通过对话安装

复制提示词发给支持本地命令或 Skills 的 AI 助手,先确认命令和权限,再让它执行。

请帮我安装这个 Agent Skill:creative-mechanics(创意机制)
来源仓库:https://github.com/motion-team/creative-strategy-skills
仓库路径:skills/creative-mechanics
安装命令:
npx skills add https://github.com/motion-team/creative-strategy-skills --skill creative-mechanics
安装前请先检查当前环境是否支持对应 CLI,并向我确认将要执行的命令、安装目录、联网范围和文件读写权限;确认后再执行。

命令行安装

复制命令到本机终端执行。该命令会通过 npx skills 从第三方来源获取 Skill;本站只展示命令,不托管安装包,也不自动执行。

skills.shnpx skills
npx skills add https://github.com/motion-team/creative-strategy-skills --skill creative-mechanics

简介

创意机制聚焦广告背后的认知与情感触发结构,而非表面形式或文案表达。

  • 适用于理解不同视觉格式下相同心理动机的复用逻辑,提升创意策略系统性。
  • 提供如“隐含答案”等核心机制解析,帮助构建可迁移的创意方法论体系。
  • 使用时应区分战略层与执行层,避免混淆机制设计与具体脚本撰写任务。
  • creative-mechanics 属于研究检索类 Skill,可作为该场景下的辅助能力补充。

SKILL.md

Creative Mechanics

Creative mechanics are the structural patterns that define how an ad creates meaning.

They are not hooks (what you say) and not visual formats (what it looks like). They are the underlying mechanism — the cognitive or emotional move the ad makes that causes the viewer to feel something, conclude something, or remember something.

The same mechanic can be executed across multiple visual formats. The Implied Answer works as a lifestyle video, a static image, a talking head, or a carousel. The mechanic travels. The format is just the vessel.


Where Mechanics Sit in the Creative Stack

Messaging Angle (Creative Strategy Engine)
    ↓
Creative Mechanic ← YOU ARE HERE
    ↓
Hook (Hook Writing + Hook Tactics)
    ↓
Visual Format

The messaging angle defines the core truth. The mechanic defines how the ad makes the viewer arrive at that truth. The hook is the opening line that triggers the mechanic. The visual format is the production structure that delivers it.


How to Use This

  1. Start with your messaging angle from the Creative Strategy Engine
  2. Ask: *how do I want the viewer to arrive at this truth?* — that's your mechanic
  3. Select the mechanic that best serves the awareness stage and persona
  4. Write the hook that opens the mechanic (Hook Writing)
  5. Choose the visual format that best executes it

The Mechanic Library


The Implied Answer

What it is: The hook poses a question that sounds like mild judgment, confusion, or curiosity. The visuals silently answer it with an aspirational truth. No one states the answer out loud — the viewer concludes it themselves.

Why it works: Self-conclusions bypass ad skepticism. When the viewer connects the dots themselves, the insight lands as their own thought, not a brand claim.

Awareness stage fit: Unaware, Problem-Aware — works best when the viewer doesn't yet know they want the product. The question meets them where they are; the visual shows them where they could be.

Structure:

  • Hook = question that sounds like an observation someone would make about the customer
  • Visual = the silent answer (the product's world, without explanation)
  • Brand never states the connection

Example:

  • Hook: *"Why do you always stay home?"*
  • Visual: Guy hanging incredible artwork all over a beautiful apartment
  • Implied truth: *His home is so good he doesn't need to leave*
  • Brand: Art/home decor

More examples by category:

  • Sleep brand → *"Why does she never want to go out anymore?"* / visual of an incredible bedroom setup
  • Home gym → *"Why does he never go to the gym?"* / visual of an insane home gym mid-workout
  • Coffee brand → *"Why does she always wake up before everyone?"* / visual of a perfect quiet morning ritual

The Social Witness

What it is: Someone other than the customer notices the change. A compliment, a double-take, an unsolicited "what are you doing differently?" The product's impact is validated through a third party's reaction — not a claim, not a review, not a before/after.

Why it works: Third-party validation is more believable than self-reported results. It also implies the change is *visible enough that other people notice* — which raises the stakes of not having the product.

Note: This is often a secondary mechanic layered onto another. The artwork ad uses Implied Answer as the primary mechanic and Social Witness as the secondary — "why do you always stay home" implies someone noticed the behavioral change.

Awareness stage fit: Problem-Aware, Product-Aware — works well when the viewer already knows they have the problem and needs to believe the product actually works.

Structure:

  • Setup: Customer living their normal life, not talking about the product
  • Witness moment: Someone else notices and comments
  • Reaction: Customer's response is low-key, not a sales pitch
  • Product appears naturally as the obvious explanation

Example:

  • Hair growth serum
  • Scene: Woman at work, colleague stops her in hallway
  • Witness: *"Wait — did you do something different? Your hair looks incredible."*
  • Reaction: She just smiles
  • Product shown at end, no voiceover

The Overheard Conversation

What it is: The ad is framed as something you weren't supposed to see — a text thread, a DM, a group chat, a conversation between friends. It feels like eavesdropping, not advertising.

Why it works: Removes the "this is an ad" filter entirely. Viewers process it as social content, not branded content. The informal framing also makes claims feel more credible — friends don't use marketing language with each other.

Awareness stage fit: Unaware, Problem-Aware — especially effective for reaching people who don't know they have the problem yet, because it mimics how people actually discover products (through friends, not ads).

Structure:

  • Frame: text thread, DM, comment section, or overheard dialogue
  • Trigger: one person mentions a problem or asks a question
  • Resolution: the other person casually recommends the product
  • No brand voiceover, no formal CTA — ends like a real conversation would

Example:

  • Gut health supplement
  • Opens on iMessage thread
  • Friend 1: *"okay don't judge me but I've been using this probiotic"*
  • Friend 2: *"omg which one"*
  • Friend 1 screenshots the product
  • Ad ends

The Reframe

What it is: Opens by validating a belief the viewer already holds — something they've been told, something they've tried, something that makes logical sense. Then flips the frame entirely. The product isn't the hero. The new perspective is.

Why it works: Viewers share things that changed how they think, not things that sold them something. The Reframe manufactures that "I never thought about it that way" moment, which drives both conversion and organic sharing.

Not to be confused with the Contrarian hook tactic. A Contrarian hook flips a belief in the opening line. The Reframe builds to the flip as the structural point of the entire concept — it's architecture, not just an opening.

Awareness stage fit: Problem-Aware, Solution-Aware — works best when the viewer has already tried things that didn't work and needs a new explanation for why.

Structure:

  • Open: validate the conventional wisdom ("you've been told X")
  • Middle: show that they've been doing everything right by that logic
  • Flip: reveal the real problem is something else entirely
  • Resolution: product solves the *real* problem, not the assumed one

Example:

  • Sleep tracking app
  • Open: *"You've been told you just need 8 hours of sleep"*
  • Middle: shows person doing everything right — in bed on time, no screens, etc.
  • Flip: *"The problem was never the hours. It was the quality."*
  • Product: the thing that finally explains why they're still exhausted

The Borrowed Enemy

What it is: The ad describes a problem, an ingredient, a feeling, or an experience that is *obviously* caused by a specific competitor — without ever naming them. The viewer makes the connection themselves.

Why it works: Naming a competitor triggers defensiveness and legal risk. Not naming them but making the reference unmistakable lets the viewer do the conquesting themselves — and again, self-conclusions land harder than stated claims.

Awareness stage fit: Solution-Aware, Product-Aware — most effective when the viewer is already shopping and comparing options, and has likely already experienced the competitor's product.

Structure:

  • Describe the problem in vivid, specific, recognizable detail
  • Use language that *only applies* to the competitor's product (specific ingredients, form factor, feeling, side effect)
  • Never name them
  • Position your product as the natural alternative

Example:

  • Clean energy drink
  • *"That jittery crash feeling... ingredients you can't pronounce... wondering what you just put in your body"*
  • Every Red Bull drinker knows exactly who this is about
  • Product presented as the obvious alternative

The Trojan Horse

What it is: The ad looks like educational content, entertainment, or a personal story — until the final 20%, where the product appears naturally as the resolution. The viewer is emotionally invested before they realize it's an ad.

Why it works: Ad avoidance is highest at the moment of recognition — "this is an ad." The Trojan Horse delays that recognition until the viewer is already engaged. By the time the product appears, it feels earned rather than inserted.

Awareness stage fit: Unaware — most powerful for cold audiences who would scroll past a traditional ad but will watch genuinely useful or entertaining content.

Structure:

  • Open as pure content: tutorial, story, observation, entertainment
  • Build genuine value or narrative tension with no product mention
  • Product enters naturally as part of the resolution — not announced, just present
  • No hard sell; the content did the work

Example:

  • Knife brand
  • Opens as a genuine cooking tutorial on how to properly dice an onion
  • Full technique breakdown, genuinely useful
  • Two minutes in: *"and obviously it helps that this knife actually holds an edge"*
  • Product was barely the point — but you just watched a two-minute ad

The Contrast Without Comment

What it is: Shows two realities side by side — before/after, with/without, old way/new way — but never editorially tells the viewer which is better. The visual makes it obvious. No voiceover passes judgment.

Why it works: Stating the obvious feels like an ad. Showing it and letting the viewer conclude feels like evidence. The absence of editorializing is itself a credibility signal.

Not to be confused with the Contrast hook tactic, which uses copy to highlight a mismatch. This mechanic removes the copy entirely and lets pure visual juxtaposition do the work.

Awareness stage fit: Problem-Aware, Solution-Aware — works well when the gap between the two states is visually dramatic enough to speak for itself.

Structure:

  • Split screen, sequential cut, or side-by-side comparison
  • Left/before: the problem state, shown neutrally
  • Right/after: the solution state, shown neutrally
  • No voiceover. No text judgment. Just the two realities.

Example:

  • Mattress brand
  • Split screen, no voiceover
  • Left: groggy wake-up, hitting snooze, dragging out of bed
  • Right: waking up naturally before the alarm, stretching, looking rested
  • Nothing said. Everything communicated.

This and a…

What it is: Two things are shown or named together — the product and something aspirational, unexpected, or emotionally resonant — with the juxtaposition doing all the work. The "and a" implies the product belongs in the same category as something the viewer already desires or respects.

Why it works: Meaning is made through association. By placing the product next to something with strong existing emotional value, the product inherits that value without claiming it directly. The viewer's brain does the association work — which lands harder than a stated claim.

Not to be confused with Contrast Without Comment, which juxtaposes two *opposing* states. "This and a…" pairs two *complementary* things to elevate the product through proximity.

Awareness stage fit: Unaware, Problem-Aware — works well for products that benefit from identity or lifestyle association rather than functional explanation.

Structure:

  • Visual or copy establishes the first thing (something aspirational, nostalgic, or culturally loaded)
  • Product is placed next to it with minimal connective tissue — just "and a" or equivalent
  • No explanation of why they belong together — the viewer feels it

Examples:

  • *"Saturdays and a Buoy in your water bottle"* — the day itself carries the vibe, the product rides it
  • *"Morning light and a quiet apartment and Loop Earplugs"* — product becomes part of a coveted aesthetic moment
  • *"A good book, an iced coffee, and a flight with Loop Earplugs"* — product slotted naturally into a fantasy travel scenario

Format note: Works especially well as a visual list or slow lifestyle montage where each element gets a beat. The rhythm of the "and a" structure is part of what makes it feel native to TikTok.


Quick Reference: Mechanic Selection

If you want the viewer to...Use this mechanic
Conclude the truth themselves from a visualImplied Answer
Believe results through someone else's reactionSocial Witness
Lower their ad guard completelyOverheard Conversation
Have a "I never thought about it that way" momentReframe
Recognize a competitor without you naming themBorrowed Enemy
Engage before they realize it's an adTrojan Horse
Feel the gap without being told what to thinkContrast Without Comment
Associate the product with something they already desireThis and a…

Layering Mechanics

Mechanics can be combined. The strongest concepts often use a primary mechanic with a secondary one reinforcing it.

Common pairings:

  • Implied Answer + Social Witness (the artwork ad — question implies someone noticed)
  • Trojan Horse + Reframe (content that builds to a perspective flip)
  • Overheard Conversation + Social Witness (the friend conversation that validates the product)
  • Contrast Without Comment + Borrowed Enemy (visual comparison that implies the competitor without naming them)

When layering, one mechanic should be primary (shapes the concept architecture) and one secondary (adds depth or a second emotional layer).


Adding New Mechanics

When a new mechanic is discovered in the wild:

  1. Describe what you saw — the hook, the visual, the structure
  2. Name the cognitive move — what did the viewer have to do to get the point?
  3. Identify why it bypassed ad resistance — what made it not feel like an ad?
  4. Tag awareness stage fit — where in the funnel does it work best?
  5. Add an example + more examples by category

Alysha adds new mechanics by describing an ad she saw and saying "add this to my creative mechanics library."

适合场景

01

用户想查找某类 Agent Skill 时

02

需要根据任务场景推荐可安装能力包时

03

需要对比不同来源的安装命令和来源信息时

能力概览

能力 1

按任务关键词查找相关 Skills

能力 2

展示可复制的安装命令

能力 3

保留来源站点、仓库和原始说明,方便继续核验

能力 4

展示第三方安全扫描或审计结果

安装后应在对应宿主中按原始 README 的触发条件使用;具体调用方式请以来源页面和 README 为准。

平台分布

Codex

38.15%
按下载量换算462

Claude

32.05%
按下载量换算388

Cursor

16.92%
按下载量换算205

Gemini CLI

10.02%
按下载量换算121

安全审计

Gen Agent Trust Hub

通过

Socket

通过

Snyk

通过

权限和风险

只读

该 Skill 主要提供规则、说明或参考内容,本身偏只读;真正读写文件、联网或执行命令仍取决于宿主 Agent 的任务。

安装前确认

本站仅展示第三方公开信息,不托管安装包,不提供自动安装或运行环境。安装前应自行审查源码、依赖和命令行为。当前只有一个来源,正式发布前建议补源仓库或其他目录站核验。

来源信息

继续浏览同类 Skills