ad-experiment
Skill1 starsChanged 59 days agoMITView on GitHub
为视频广告设计素材对照实验,分析已有投放数据与创意变量。用于开头、正文或 CTA 测试和效果复盘;不把观看感受、样片质量或小样本领先当成转化证明。
In this repository10 other files it gives its agents.
- ad-briefSkill
- ad-creativeSkill
- ad-discoverySkill
- ad-gtmSkill
- ad-pocSkill
- ad-reviewSkill
- ad-scaleSkill
- ad-seedance-20Skill
- ad-seedance-25Skill
- ad-storyboardSkill
The file
Files
素材实验与复盘
输入
实验目标、候选素材、改变项、渠道、主要指标、可用预算与周期。结果分析另需曝光、点击、转化、花费、归因窗口和流量分配等实际记录。
方法
- 写一个可检验假设,指定主要指标与停止规则。指标定义使用实际渠道口径;没有基线先做探索,不事后挑选有利指标。
- 分配
experiment_id / variant_id / asset_id / ad_id,记录提示词或剪辑版本。优先一次改变一项,固定正文、商品事实、CTA 等其余部分。 - 明确受众、广告位、出价、时段和归因窗口如何保持可比;记录随机分流或平台自动优化机制。不能随机分配时说明结果是观察性比较。
- 在执行前确定样本或预算、周期、最小有用效果和分析方法。用户没有提供转化基线时,不编造固定样本量或显著性阈值。
- 有授权和工具才创建投放或花费;否则只交付实验计划。分析已有数据时先检查缺失、重复、单位及分母。
- 报告原始计数、比率、变化幅度与不确定性。CTR = 点击 / 曝光;CVR 要注明按点击还是其他事件作分母;CPA = 花费 / 转化。分母为零时不计算对应指标。
- 对照预先约定的方法判断继续、停止或补数据。区分业务上有用的变化、统计证据与可能的混杂因素。
输出
实验卡:假设、变体、固定项、主要指标、分流、预算、周期、归因、停止与分析规则。
结果卡:各组计数和指标、计算口径、是否满足样本要求、限制、结论与下一轮只改变什么。没有数据时保持 not_run。
完成标准与交接
每项效果判断都有真实数据及口径;样本不足时明确“不足以判断”。将观察到的创意问题回传 ad-creative,客户商业结论可传给 ad-poc 或 ad-scale,不混淆制作与媒体成本。
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
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.

