Token导航 LogoToken导航

Agent Skill · 研究检索

ad-creative(广告创意)

ad-creative 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。

ad-creative 是研究检索类 Agent Skill,可用宿主为 Codex、Claude、Cursor、Gemini CLI,通过GitHub 仓库获取,权限为需要联网;许可证信息还不明确,第三方扫描已通过,按安装量在研究检索类的 73,675 个 Skill 里排第 36,169。

总安装 1,972周安装 79GitHub Stars 103下载 638核验 2026-05-01

安装说明

本站只整理中文说明,不托管安装包,也不代用户安装。
GitHub
许可证
unknown

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:ad-creative(广告创意)
来源仓库:https://github.com/borghei/claude-skills
仓库路径:skills/ad-creative
安装命令:
npx skills add https://github.com/borghei/claude-skills --skill ad-creative
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/borghei/claude-skills --skill ad-creative

这个 Skill 是做什么的

研究检索类 Skill 主要解决「找资料、核对事实、把来源整理清楚」的问题,通常会让助手先检索再总结。

专注于跨平台广告创意设计与优化,支持多平台规格适配和素材迭代。适用于 Meta、Google、TikTok 等主流平台的广告文案、图片及视频创意生成。提供关键词研究、A/B 测试框架和效果评估方法,提升广告素材表现力。使用前请确认目标受众与平台政策,避免违反广告规范或版权要求。

ad-creative 属于研究检索类 Skill,可作为该场景下的辅助能力补充。

来源仓库内路径:skills/ad-creative

上手步骤

4 步
  1. 1

    确认自己的宿主工具

    公开资料标注支持 Codex、Claude、Cursor、Gemini CLI;先确认自己用的是其中之一,再按对应工具的 Skills 目录放置。

  2. 2

    拿到安装命令

    可以直接复制的命令是 npx skills add https://github.com/borghei/claude-skills --skill ad-creative;命令里的占位符换成自己的路径或密钥再执行。

  3. 3

    安装后重启宿主

    命令执行完要完全退出宿主工具再打开,让新装的 Skill 被重新扫描到,否则列表里可能看不到。

  4. 4

    按最小权限试用

    它需要需要联网,先在测试目录或测试账号里跑,确认没问题再放到正式项目。

常见问题

6 条

ad-creative 怎么安装?

在终端执行 npx skills add https://github.com/borghei/claude-skills --skill ad-creative。装完完全退出宿主工具再打开,让它重新扫描本地 Skill。

ad-creative 支持哪些 AI 工具?

公开资料标注支持 Codex、Claude、Cursor、Gemini CLI。其他工具只要能读取本地 Skill 目录,一般也能手动放进去,但要自己确认触发方式是否一致。

ad-creative 需要什么权限?

权限级别是需要联网。该 Skill 可能需要联网访问来源站点、仓库或外部 API;具体网络访问范围需要结合源码和 README 复核。

ad-creative 是免费的吗?

公开资料没有标注明确许可证,收费和商用限制都无法确认。正式使用前请到来源仓库核对授权说明。

ad-creative 用的人多吗?

公开数据是累计安装 1,972 次,近一周安装 79 次,下载 638 次,GitHub 103 星。这些数字来自第三方目录统计,只能当作热度参考,不代表质量或安全性。

ad-creative 安全吗?

第三方扫描结果为通过,但这只覆盖已知的静态检查项,不能替代自己读一遍源码和安装命令。

值得选的地方和要先想清楚的地方

值得选的地方

  • 同时适配 4 种工具,换客户端不用重新找替代品。
  • 第三方扫描已通过,常见的高风险行为没有命中。

先想清楚的地方

  • 许可证没有明确标注,能不能商用、能不能再分发都不确定。
  • 第三方目录整理,版本更新和行为细节以来源仓库为准。

适不适合你

更适合

  • 已经在用 Codex,想直接扩展它能力的人。
  • 需要处理研究检索类任务、又不想每次都手写提示词的人。

不太适合

  • 需要明确授权才能用到公司项目里的场景。
  • 需要厂商级技术支持和 SLA 的生产系统。

研究检索类 Skill 对比

6 个同类
Skill总安装GitHub 星权限许可证扫描
find-skills

寻找技能 · Codex、Claude

31,199,81316,486操作浏览器MIT有提醒
self-improving-agent

自我改进 Agent · OpenClaw

16,464,0023,396接触敏感数据MIT-0通过
soultrace

灵魂追踪 · Codex、Claude

7,588,94412需要联网未明确有提醒
self-improving

自我完善 · OpenClaw

6,829,2841,046需要联网MIT-0通过
microsoft-foundry

微软代工厂 · Codex、Claude

6,614,144764调用外部服务未明确有提醒
ad-creative

广告创意 · 本页

1,972103需要联网未明确通过

按安装量取同分类前几名;数字来自第三方目录统计,只作热度参考,不代表质量结论。

SKILL.md 原文

Ad Creative

Production-grade ad creative design, iteration, and optimization across all major advertising platforms.


Table of Contents


Keywords

ad creative, ad copy, headline generation, RSA headlines, Meta ad copy, LinkedIn ads, Google Ads, Twitter/X ads, TikTok ads, creative testing, A/B testing, ad variations, bulk creative, performance creative, display ads, search ads, social ads, CTA optimization, ad compliance, character limits, creative matrix, ad iteration, conversion copy, platform-specific ads


Quick Start

Generate Ad Copy from Scratch

  1. Define product, audience, and funnel stage
  2. Select platform and ad format from the specs table
  3. Choose creative framework matching funnel stage
  4. Generate 8-15 headline variations using formula library
  5. Write body copy per platform character limits
  6. Run quality validation checklist before submission

Iterate on Existing Ads

  1. Collect performance data (CTR, CVR, CPA) for current ads
  2. Diagnose winning pattern: hook type, emotional driver, CTA style
  3. Generate 3-5 on-theme variations preserving the winning pattern
  4. Generate 2-3 new angle tests for exploration
  5. Validate all copy against platform compliance rules

Core Workflows

Workflow 1: Full Creative Set Generation

Step 1: Define the Creative Brief

Document before writing any copy:

## Creative Brief
- Product/Offer: [Specific product, feature, or lead magnet]
- Value Proposition: [One sentence — what changes in the customer's life]
- Target Audience: [Job title, pain point, awareness level]
- Platform(s): [Google / Meta / LinkedIn / Twitter/X / TikTok]
- Funnel Stage: [Awareness / Consideration / Decision]
- Existing Creative: [Yes/No — if yes, attach performance data]
- Budget Context: [Daily/monthly spend, CPA targets]

Step 2: Generate Headlines by Formula Type

Produce minimum 10 headlines, distributed across formula categories:

  • 3 benefit-first headlines
  • 2 curiosity headlines
  • 2 social proof headlines
  • 2 problem agitation headlines
  • 1 urgency headline (only if genuine scarcity exists)

Step 3: Write Platform-Specific Body Copy

For each platform, write 2-3 body copy variations:

  • Respect character limits exactly
  • Match tone to platform norms
  • Include CTA aligned with landing page offer

Step 4: Assemble Ad Sets

Organize into testable ad sets:

[AD SET: Benefit-First] | [Platform] | [Funnel Stage]
  Headline A: "..." (28/30 chars)
  Headline B: "..." (26/30 chars)
  Body: "..." (87/90 chars)
  CTA: "Start Free Trial"

Step 5: Validate and Submit

Run every ad through the quality checklist before upload.

Workflow 2: Performance-Based Iteration

Step 1: Audit Current Creative

For each running ad, classify:

  • Hook type: Problem / Benefit / Curiosity / Social Proof / Urgency
  • Emotional driver: Fear / Ambition / FOMO / Frustration / Relief
  • CTA type: Click / Sign up / Learn more / Book / Download
  • Performance tier: Winner / Average / Underperformer

Step 2: Extract the Winning Pattern

Identify what the top performers share:

  • Specific numbers vs. vague claims
  • Customer voice vs. brand voice
  • Direct benefit vs. emotional appeal
  • Short vs. long format

Step 3: Generate Variations

  • 3-5 on-theme variations (same hook type, different angle)
  • 2-3 exploration variations (new hook type, untested territory)

Step 4: Build a Creative Testing Matrix

AngleGoogleMetaLinkedIn
Benefit-first[copy][copy][copy]
Social proof[copy][copy][copy]
Problem agitation[copy][copy][copy]

Workflow 3: Cross-Platform Adaptation

Step 1: Lock the Core Message

Define the single transferable message that works across all platforms.

Step 2: Adapt Per Platform

Reformat (not rewrite from scratch):

  • Adjust character count to platform specs
  • Modify tone (LinkedIn = professional, TikTok = casual, Google = direct)
  • Change CTA style to match platform conventions
  • Adjust visual/copy balance per platform norms

Platform Specifications

Character Limits Reference

PlatformFormatHeadline LimitBody/Description LimitKey Constraints
Google RSASearch30 chars x 1590 chars x 4 descriptionsMax 3 pinned positions
Google DisplayDisplay30 chars x 5 short, 90 chars x 1 long90 chars x 5Requires 5+ images
Google Performance MaxMulti30 chars x 5 short, 90 chars x 5 long90 chars x 5Auto-placement
Meta FeedImage/Video40 chars headline125 chars primary text (recommended)Image text < 20%
Meta StoriesVertical40 chars125 chars9:16 aspect ratio
LinkedIn SponsoredContent70 chars headline150 chars intro textNo clickbait policies
LinkedIn MessageInMail60 chars subject1,500 chars bodyPersonalization required
Twitter/XPromoted70 chars headline280 chars totalNo deceptive tactics
TikTok In-FeedVideoNo text overlay limit80-100 chars captionHook in first 2-3 seconds
TikTok SparkNativeVaries100 charsMust use creator content

Platform-Specific Rejection Triggers

Google Ads:

  • ALL CAPS (except acronyms like SaaS, API)
  • Excessive punctuation (!!!,???,....)
  • Misleading claims ("guaranteed #1 ranking")
  • Trademarked competitor names in headlines
  • Gimmicky formatting (s.p.a.c.e.d letters)

Meta Ads:

  • Image text exceeding 20% of image area
  • Before/after body transformation images
  • Personal attributes assumptions ("Are you overweight?")
  • Fake UI elements (play buttons, notifications)
  • Sensationalized content

LinkedIn Ads:

  • Clickbait headlines
  • Misleading claims about job opportunities
  • Profanity or inappropriate language
  • Targeting by sensitive categories

Creative Frameworks by Funnel Stage

Awareness Stage: Lead with the Problem

The audience does not know you. Meet them where they are.

Framework: Problem > Amplify > Hint

Hook: [State the problem in their language]
Amplify: [Show the cost of inaction]
Hint: [Tease that a solution exists — don't pitch yet]

What works at awareness:

  • Curiosity hooks that create an open loop
  • Statistic-led hooks with surprising data
  • "You know that feeling when..." relatable hooks
  • Questions that make them stop scrolling

What fails at awareness:

  • Product pitches to cold audiences
  • Feature lists nobody asked for
  • Brand-centric messaging ("We're the leading...")

Consideration Stage: Lead with the Solution

They know the problem. They are evaluating options.

Framework: Solution > Mechanism > Proof

Hook: [Name the outcome they want]
Mechanism: [Explain how you deliver it differently]
Proof: [Show evidence it works]

What works at consideration:

  • Benefit-first headlines with specificity
  • Comparison frames ("Unlike X, we do Y")
  • How-it-works explanations in 3 steps
  • Case study references with real numbers

Decision Stage: Lead with Proof

They are close. Remove the last objection.

Framework: Proof > Risk Removal > Action

Hook: [Testimonial, case study, or result with numbers]
Risk Removal: [Free trial, money-back, no credit card]
Action: [Specific, clear CTA]

What works at decision:

  • Social proof headlines with specific metrics
  • Guarantee-first positioning
  • Before/after transformation stories
  • Urgency only if real (limited slots, deadline-based pricing)

Headline Formula Library

Benefit-First Formulas

[Verb] [specific outcome] [timeframe or qualifier]

Examples:

  • "Cut your churn rate by 30% without chasing customers"
  • "Ship features your team actually uses"
  • "Hire senior engineers in 2 weeks, not 4 months"
  • "Reduce support tickets by 60% this quarter"

Curiosity Formulas

[Surprising claim or counterintuitive angle]

Examples:

  • "The email sequence that gets replies when your first one fails"
  • "Why your best customers leave at 90 days"
  • "What your analytics dashboard is hiding from you"

Social Proof Formulas

[Number] [people/companies] [outcome]

Examples:

  • "1,200 SaaS teams use this to reduce support tickets"
  • "Trusted by 40,000 developers across 80 countries"
  • "How Stripe doubled activation in 6 weeks"

Problem Agitation Formulas

[Describe the pain vividly in their language]

Examples:

  • "Still losing 40% of signups before they see value?"
  • "Your ads are running, your budget is spending, and you can't tell what's working"
  • "Another month of manually updating spreadsheets?"

Urgency Formulas (Use Only with Genuine Scarcity)

[Real constraint + specific deadline or quantity]

Examples:

  • "Q1 pricing ends March 31 — new rates from April 1"
  • "Only 3 onboarding slots open this month"

Anti-pattern to avoid:

  • "LIMITED TIME DEAL!! ACT NOW!!!" — gets rejected and destroys trust

Iteration Methodology

Step 1: Diagnose the Winner

For every winning ad, document:

DimensionAnswer
Hook typeProblem / Benefit / Curiosity / Social Proof
Funnel stage servedAwareness / Consideration / Decision
Emotional driverFear / Ambition / FOMO / Frustration / Relief
CTA friction levelLow (learn more) / Medium (free trial) / High (book demo)
Specificity levelVague claims / Specific numbers / Case study reference

Step 2: Generate On-Theme Variations

Preserve the winning pattern, vary one element at a time:

  • Same hook type, different product angle
  • Same emotional driver, different customer scenario
  • Same structure, different proof point

Step 3: Test New Territory

For every 3-5 on-theme variations, include 2-3 exploration ads:

  • Different hook type than the winner
  • Different emotional driver
  • Different audience segment language

Step 4: Measure and Learn

After 7-14 days with statistical significance:

  • Promote winners to main rotation
  • Pause underperformers
  • Document what worked for the creative playbook
  • Feed learnings into next iteration cycle

A/B Testing Framework

Testing Hierarchy (Highest to Lowest Impact)

  1. Creative concept/angle — The biggest lever. Test different value propositions
  2. Hook/headline — First impression. Test different opening patterns
  3. Visual style — Image vs. video, lifestyle vs. product, dark vs. light
  4. Body copy — Supporting argument. Test different proof points
  5. CTA — Action text. Test different friction levels

Testing Rules

  • Test one variable at a time per ad set
  • Minimum 1,000 impressions per variant before drawing conclusions
  • Run tests for 7-14 days minimum
  • Use platform-native A/B tools when available
  • Document every test hypothesis, variant, and result
  • Calculate statistical significance before declaring winners (95% confidence)

Test Documentation Template

## A/B Test: [Name]
- Hypothesis: [If we change X, then Y will improve because Z]
- Variable: [Headline / Body / CTA / Visual]
- Control: [Current version]
- Variant: [New version]
- Metric: [CTR / CVR / CPA / ROAS]
- Duration: [Start date - End date]
- Sample size: [Impressions per variant]
- Result: [Winner + confidence level]
- Learning: [What we learned for future creative]

Quality Validation

Pre-Submission Checklist

Platform Compliance:

  • All character counts within platform limits
  • No ALL CAPS except standard acronyms
  • No excessive punctuation (!!! or??? or....)
  • No trademarked competitor names in restricted positions
  • No platform name references in copy ("Facebook," "Google")
  • No fake UI elements in images
  • Image text ratio under 20% for Meta

Copy Quality:

  • Headline makes sense standalone without description
  • Specific claims over vague claims ("save 3 hours" not "save time")
  • CTA matches the landing page offer exactly
  • No unsubstantiated superlatives (#1, best-in-class, industry-leading)
  • No claims that require disclaimers you have not included

Strategic Alignment:

  • Copy matches the target funnel stage
  • Language matches how the audience describes this problem
  • Hook type varies across the ad set (not all the same formula)
  • Landing page message matches ad promise

Common Rejection Reasons to Pre-Check:

  • Misleading destination (ad says X, landing page says Y)
  • Unsubstantiated claims about results or performance
  • Trademark violations in headline text
  • Policy-violating content categories (alcohol, healthcare claims, financial guarantees)

Best Practices

  1. Write 3x more headlines than you need — Generate 15-20, select the best 8-10. Creative quality increases with volume.
  2. Pin headlines strategically — In Google RSA, pin only your strongest headline to Position 1. Let the algorithm optimize the rest.
  3. Never reuse landing page copy verbatim — Ad copy and landing page should feel connected but not identical. The ad earns the click; the page earns the conversion.
  4. Refresh creative every 2-4 weeks — Ad fatigue degrades performance. Track frequency metrics and replace creative before CTR declines.
  5. Localize, do not just translate — Different markets require different cultural references, humor styles, and proof points.
  6. Match CTA friction to funnel stage — Awareness: "Learn more." Consideration: "See how it works." Decision: "Start free trial."
  7. Use dynamic keyword insertion carefully — It improves relevance but can produce awkward headlines. Always preview all combinations.
  8. Test video vs. static on Meta and TikTok — Video consistently outperforms static for awareness, but static can win for retargeting.
  9. Lead with mobile — 80%+ of social ad impressions are mobile. Design for small screens first.
  10. Keep proof points current — "Trusted by 500 companies" should update as you grow. Stale numbers erode credibility.

Integration Points

  • Paid Ads — Use for campaign strategy, audience targeting, and budget optimization. Ad Creative handles the copy; Paid Ads handles the campaign.
  • Copywriting — Use for landing page and long-form web copy. Ad creative follows different constraints (character limits, platform compliance).
  • Campaign Analytics — Use to measure ad performance and feed learnings back into the iteration cycle.
  • Marketing Psychology — Use psychological principles (anchoring, social proof, loss aversion) to strengthen ad messaging.
  • Brand Guidelines — Reference brand voice and visual standards to maintain consistency across ad creative.

Troubleshooting

SymptomLikely CauseFix
Ad rejected by platformPolicy violation (caps, claims, trademarked terms)Run headline_generator.py to generate compliant variants. Check rejection triggers.
All headlines same styleNo formula diversityGenerate across 5 formula types: benefit, curiosity, social proof, problem, urgency.
CTR declining across ad setCreative fatigueRun creative_fatigue_detector.py. Refresh every 2-4 weeks.
High CTR but low conversionsAd-to-landing-page mismatchEnsure headline promise matches landing page. Message match is the #1 conversion factor.
Can't fill all Google RSA slotsNot enough headline variationsUse headline_generator.py to produce 15+ variations from value proposition.
Video underperforming on MetaHook not in first 2-3 seconds80%+ of social impressions are mobile. Hook must appear in first 2 seconds.
Same CPA across all creativeNot testing enough anglesBuild a creative testing matrix with 5 hook types x platforms. Use creative_matrix_builder.py.

Success Criteria

  • 3+ creative variants per ad group/set (minimum for testing)
  • All copy within platform character limits with zero rejections
  • Headline formula mix: benefit, curiosity, social proof, problem, urgency
  • Creative refreshed every 2-4 weeks before fatigue signals appear
  • A/B test documentation for every test: hypothesis, variants, result, learning
  • Winning patterns documented in creative playbook for institutional knowledge
  • CTR above platform benchmarks with declining CPA trend

Scope & Limitations

In Scope: Ad headline/body copy generation, platform-specific formatting, A/B testing frameworks, creative iteration methodology, compliance validation, cross-platform adaptation.

Out of Scope: Visual design/production, video editing, campaign management (use paid-ads), landing page copy (use copywriting), media buying.


Python Automation Tools

1. Headline Generator (scripts/headline_generator.py)

Generates ad headline variations from a value proposition using proven formula patterns, scored for platform compliance.

python scripts/headline_generator.py --value-prop "Reduce churn by 30%" --platform google --audience "SaaS teams"
python scripts/headline_generator.py --value-prop "Reduce churn by 30%" --json

2. Creative Fatigue Detector (scripts/creative_fatigue_detector.py)

Analyzes ad performance trends (CTR, CPA, frequency) to detect creative fatigue and recommend refresh timing.

python scripts/creative_fatigue_detector.py campaign_data.json
python scripts/creative_fatigue_detector.py --sample --json

3. Creative Matrix Builder (scripts/creative_matrix_builder.py)

Generates a cross-platform creative testing matrix from a brief, organized by hook type and funnel stage.

python scripts/creative_matrix_builder.py brief.json
python scripts/creative_matrix_builder.py --sample --json

平台分布

Codex

33.57%
按下载量换算214

Claude

31.03%
按下载量换算198

Cursor

16.66%
按下载量换算106

Gemini CLI

9.82%
按下载量换算63

安全审计

Gen Agent Trust Hub

通过

Socket

通过

Snyk

通过

权限和风险

需要联网

该 Skill 可能需要联网访问来源站点、仓库或外部 API;具体网络访问范围需要结合源码和 README 复核。

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。目前只有一个公开来源,建议同时核对源仓库。

继续浏览同类 Skills