Token导航 LogoToken导航

Agent Skill · 开发

tiktok-ads(抖音广告)

tiktok-ads 用于记录任务执行中的错误、用户纠正、经验和能力缺口,适合在 OpenClaw 中希望让 Agent 持续沉淀问题、修正和最佳实践时使用。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。

tiktok-ads 是开发类 Agent Skill,可用宿主为 OpenClaw,通过ClawHub获取,权限为需要联网;许可证为 MIT-0,第三方扫描已通过,按安装量在开发类的 21,746 个 Skill 里排第 480。

总安装 86,577周安装 3,681GitHub Stars 12下载 30,331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
MIT-0
版本
1.0.0

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:tiktok-ads(抖音广告)
来源仓库:https://github.com/ivangdavila/tiktok-ads
安装命令:
openclaw skills install tiktok-ads
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 OpenClaw 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。

ClawHubOpenClaw
openclaw skills install tiktok-ads

这个 Skill 是做什么的

开发类 Skill 面向写代码、改代码和排查问题,一般会读写仓库文件,部分还会执行构建或测试命令。

tiktok-ads 提供 TikTok 广告策略与创意最佳实践指导。适用于 OpenClaw 中制定 Spark Ads 投放计划或优化直播带货素材。包含平台特有规则、出价模型与受众定向建议,辅助提升 ROI。

安装命令为 openclaw skills install tiktok-ads,需接入 TikTok 广告账户权限。注意广告政策变动,避免因违规导致预算冻结或账户处罚。

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 OpenClaw;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 openclaw skills install tiktok-ads;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

tiktok-ads 怎么安装?

在终端执行 openclaw skills install tiktok-ads。装完完全退出宿主工具再打开,让它重新扫描本地 Skill。

tiktok-ads 支持哪些 AI 工具?

公开资料标注支持 OpenClaw。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。

tiktok-ads 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。

tiktok-ads 是免费的吗?

来源标注的许可证是 MIT-0。本站不代收任何费用,也不托管安装包;能否商用、要不要署名,以许可证原文为准。

tiktok-ads 用的人多吗?

公开数据是累计安装 86,577 次,近一周安装 3,681 次,下载 30,331 次,GitHub 12 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。

tiktok-ads 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

  • 专注 OpenClaw 一种宿主,触发方式和使用习惯不用来回切换。
  • 许可证明确为 MIT-0,商用和分发边界好判断。
  • 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

  • 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 OpenClaw,想直接扩展它能力的人。
  • 需要处理开发类任务、又不想每次都手写提示词的人。
  • 需要许可证清楚、并且有第三方扫描结果的团队项目。

不太适合

  • 需要厂商级技术支持和 SLA 的生产系统。

开发类 Skill 对比

6 个同类
Skill总安装GitHub 星权限许可证扫描
skill-vetter

技能审查员 · OpenClaw

8,779,018986操作浏览器MIT-0通过
ontology

本体论 · OpenClaw

6,636,960576只读MIT-0通过
entra-app-registration

内部应用程序注册 · Codex、Claude

6,631,872764接触敏感数据未明确通过
entra-app-registration

内部应用程序注册 · Codex、Claude

3,757,680201接触敏感数据未明确通过
nano-banana-pro

纳米香蕉专业版 · OpenClaw

3,526,898364接触敏感数据MIT-0有提醒
tiktok-ads

抖音广告 · 本页

86,57712需要联网MIT-0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

name
TikTok Ads
description
TikTok advertising strategy, creative best practices, Spark Ads, and platform-specific optimization
metadata
category
marketing
skills
["tiktok", "ads", "social-ads", "video-ads", "short-form"]

Creative Rules

  • First 1-3 seconds determine everything — if viewers don't stop scrolling immediately, the rest doesn't matter
  • Native content outperforms ads — shoot vertical (9:16), use trending sounds, avoid polished brand look
  • Hook patterns that work: pattern interrupt, bold claim, direct question, unexpected visual, POV setup
  • Text overlays are mandatory — 80%+ of users watch with sound off, burned-in captions aren't optional
  • Refresh creative every 7-14 days — fatigue hits faster than any other platform, have pipeline ready

Spark Ads

  • Spark Ads use organic posts as ads — keeps comments, likes, shares visible and accumulating
  • Organic engagement on Spark Ads compounds — viral potential unlike standard ads
  • Creator handles can be swapped — run ads from your brand handle or creator's handle
  • Require creator authorization code — valid for 7, 30, or 365 days, get longer duration upfront
  • Boosting already-performing organic posts is safer than untested creative — data before dollars

Targeting Mistakes

  • Interest targeting is less accurate than Meta — TikTok's algorithm learns faster from creative signals
  • Start broad, let algorithm optimize — overtargeting restricts the machine learning that makes TikTok work
  • Custom Audiences need 1000+ users minimum — smaller lists won't deliver
  • Lookalike percentages are different scale — 1% on TikTok is broader than 1% on Meta
  • Behavior targeting (video completion, engagement) often outperforms interest targeting

Bidding Strategy

  • Learning phase needs 50 conversions in 7 days — budget must support this or campaign never optimizes
  • Cost Cap is stabler than Bid Cap for beginners — limits volatility while algorithm learns
  • Lowest Cost burns budget fast on launch — switch to Cost Cap after gathering initial data
  • Minimum daily budget per ad group: $20-50 depending on objective — below this, delivery is inconsistent
  • Don't edit campaigns during learning phase — resets progress, wait until 50 conversions

Campaign Structure

  • One campaign per objective — mixing awareness and conversion in same campaign confuses optimization
  • 3-5 ad groups per campaign is optimal — enough for testing, not so many that budget fragments
  • 3-5 creatives per ad group — algorithm needs options to test but too many dilutes learnings
  • Separate iOS and Android campaigns post-ATT — iOS tracking is limited, don't let it pollute Android data
  • Catalog ads for e-commerce — dynamic product ads outperform static for retargeting

Platform Mechanics

  • Content is served via For You algorithm, not follower graph — follower count matters less than Meta
  • Hashtag challenges are paid products — organic hashtags have minimal targeting value in ads
  • Comments section is public and active — negative comments kill CTR, moderate or disable if problematic
  • Video completion rate is primary signal — optimize for watch time, not just clicks
  • TikTok pixel has same iOS limitations as Meta — use Events API for server-side tracking

Creative Testing

  • Test hooks independently from body — first 3 seconds are separate variable from rest of video
  • A/B test in creative tools, not ad platform — TikTok's native split testing is limited
  • User-generated style beats studio quality — imperfect > polished on this platform
  • Test creator vs brand voice — same script performed differently converts at different rates
  • Winners emerge fast — if no signal in 3-5 days, creative likely won't work

Conversion Tracking

  • Standard pixel misses 40%+ of iOS conversions — implement Events API for accuracy
  • View-through window defaults to 7 days — conservative window (1 day) shows truer performance
  • Lead gen forms have high drop-off — keep fields to absolute minimum (name, email, phone)
  • TikTok Shop has native tracking — if selling via Shop, conversion data is more complete
  • Attribution is last-click by default — view-through conversions are modeled, not deterministic

Scaling Patterns

  • Increase budget 20-30% max per day — larger jumps reset learning phase
  • Horizontal scaling: duplicate ad groups with different targeting — avoids audience overlap issues
  • Creative is the scaling lever — same audience with new creative > same creative with forced reach
  • CPMs spike during holidays and elections — plan 2-3x budget for Q4 and avoid political seasons if possible
  • International markets have lower CPMs — test UK, Canada, Australia before saturating US

Common Failures

  • Repurposing Instagram Reels directly — TikTok users detect non-native content instantly
  • Logo in first 3 seconds — viewers skip ads, not content. Logo at end, not beginning
  • Long intros or brand statements — no one waits, hook must be immediate
  • Ignoring comments — negative comments are social proof against you, moderate actively
  • Expecting Meta-level tracking accuracy — TikTok attribution is less mature, expect discrepancies

平台分布

OpenClaw

95.35%
按下载量换算28,921

安全审计

VirusTotal

通过

ClawScan

通过

Static analysis

未展示

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