规划页面改动的A/B测试,从提出假设、确定指标,到估算样本量、测试时长和流量分配,并帮助解读结果、排序实验方案及记录持续改进中的经验。
- 比较标题或注册按钮
- 估算样本量和测试时长
- 检查结果及可能的不利影响
整理产品特点、目标客户、竞争差异、客户用语和品牌语气,维护共享营销资料,供其他营销技能参考,并记录版本变化与实践经验。
When the user wants to create or update their product marketing context document. Also use when the user mentions 'product context,' 'marketing context,' 'set up context,' 'positioning,' 'who is my target audience,' 'describe my product,' 'ICP,' 'ideal customer profile,' 'brand voice,' 'tone of voice,' 'brand guidelines,' 'what have we learned,' 'log this learning,' or wants to avoid repeating foundational information across marketing tasks. Use this at the start of any new project before using other marketing skills — it creates `.agents/product-marketing.md` that all other skills reference for product, audience, and positioning context.
实测评分 未实测
查看安装方法本技能会将产品特点、目标客户、与竞争对手的差异、客户用语和品牌语气整理到.agents/product-marketing.md。其他营销技能可以参考这份共享资料,减少每次任务都重新说明基础信息的需要。资料更新时会记录版本号和修改原因,也可以管理营销活动中积累的经验。
开始新项目的营销工作之前,或需要统一广告、邮件和社交媒体文案的宣传重点时,可以使用本技能。可以根据README或现有产品介绍页面等资料起草,也可以通过对话逐项补充。目标客户或产品定位发生变化时,也可以用它检查和更新共享信息。
初稿和更新内容都需要使用者检查。新增营销经验时,只有在使用者批准后才会保存。涉及多人参与采购的人物画像适用于B2B;广告相关项目只在实际投放广告时处理。
本介绍由 AI(OpenAI Codex)根据公开的 SKILL.md 撰写。准确的规格请查看下方原文。
AI 根据 SKILL.md 的内容检查了需要注意的操作和使用前提。评分只依据实际运行的结果(实测对比)。
不计入评分。 由 AI(OpenAI Codex)于 2026年10月24日 评估。仅供参考,安装前请同时查看原文。
安装前可以查看交给 AI 代理的指令内容。
You help users create and maintain a product marketing context document. This captures foundational positioning and messaging information that other marketing skills reference, so users don't repeat themselves.
The document is stored at .agents/product-marketing.md.
First, check if .agents/product-marketing.md already exists. Also check .claude/product-marketing.md and the legacy filename product-marketing-context.md (in either .agents/ or .claude/) for older setups — if found anywhere other than .agents/product-marketing.md, offer to move it to the canonical location.
If it exists:
If it doesn't exist, offer two options:
Auto-draft from codebase (recommended): You'll study the repo—README, landing pages, marketing copy, package.json, etc.—and draft a V1 of the context document. The user then reviews, corrects, and fills gaps. This is faster than starting from scratch.
Start from scratch: Walk through each section conversationally, gathering info one section at a time.
Most users prefer option 1. After presenting the draft, ask: "What needs correcting? What's missing?"
If auto-drafting:
If starting from scratch: Walk through each section below conversationally, one at a time. Don't dump all questions at once.
For each section:
Push for verbatim customer language — exact phrases are more valuable than polished descriptions because they reflect how customers actually think and speak, which makes copy more resonant.
If multiple stakeholders are involved in buying, capture for each:
The JTBD Four Forces:
Build this from copy that already performed and from the founder's own writing, rather than from adjectives picked in a meeting. See references/brand-voice.md for the extraction steps, a worked profile, before/after examples, and per-person voice guides.
The ads skill reads this so the user doesn't re-explain the account every session.
Start this empty. It fills up as work ships (see "Maintaining Learnings" below).
After gathering information, create .agents/product-marketing.md with this structure:
# Product Marketing Context
**Document version:** v1
**Last updated:** [date]
## Product Overview
**One-liner:**
**What it does:**
**Product category:**
**Product type:**
**Business model:**
## Target Audience
**Target companies:**
**Decision-makers:**
**Primary use case:**
**Jobs to be done:**
-
**Use cases:**
-
## Personas
| Persona | Cares about | Challenge | Value we promise |
|---------|-------------|-----------|------------------|
| | | | |
## Problems & Pain Points
**Core problem:**
**Why alternatives fall short:**
-
**What it costs them:**
**Emotional tension:**
## Competitive Landscape
**Direct:** [Competitor] — falls short because...
**Secondary:** [Approach] — falls short because...
**Indirect:** [Alternative] — falls short because...
## Differentiation
**Key differentiators:**
-
**How we do it differently:**
**Why that's better:**
**Why customers choose us:**
## Objections
| Objection | Response |
|-----------|----------|
| | |
**Anti-persona:**
## Switching Dynamics
**Push:**
**Pull:**
**Habit:**
**Anxiety:**
## Customer Language
**How they describe the problem:**
- "[verbatim]"
**How they describe us:**
- "[verbatim]"
**Words to use:**
**Words to avoid:**
**Glossary:**
| Term | Meaning |
|------|---------|
| | |
## Brand Voice
| Attribute | Means | Doesn't mean | Example line |
|-----------|-------|--------------|--------------|
| | | | |
**Tone by context:**
- Support:
- Launch:
- Errors:
- Sales:
- Social:
**Terms we own:**
**Never use (and why):**
## Proof Points
**Metrics:**
**Customers:**
**Testimonials:**
> "[quote]" — [who]
**Value themes:**
| Theme | Proof |
|-------|-------|
| | |
## Goals
**Business goal:**
**Conversion action:**
**Current metrics:**
## Paid Ads (optional)
**Target CPA / ROAS:**
**Hard floor:**
**Monthly budget and daily variance:**
**Conversions that count:**
| Action | Means | Value | Feeds bidding? |
|--------|-------|-------|----------------|
| | | | |
**Known tracking weaknesses:**
**Good lead vs. form fill:**
**Campaigns:** [name] — [what it's for]
**Agent may:** read, compare, flag, draft, recommend
**Agent may not (without approval):** change budgets or bids, pause or launch, edit tracking, publish ads
## Learnings
*Decisions and tested lessons. Newest first. Keep it to about 20 active entries.*
| Date | Learning | Evidence | Applies to | Status |
|------|----------|----------|------------|--------|
| | | | | |
## Changelog
*Newest first. One line per revision: what changed and why.*
- v1 ([date]) — Initial context.
Document version: v1 and a single Changelog entry — - v1 ([today]) — Initial context.Last updated to today, and prepend a new Changelog entry at the top of the list (newest first) summarizing what changed and why in one line. Never rewrite or reorder past entries.- v3 (2026-07-16) — Repositioned from "email tool" to "deliverability platform"; added RevOps to the ICP.- v2 (2026-06-02) — Rewrote value prop and objections after 5 customer interviews; added competitor Acme.Last updated..agents/product-marketing.md/product-marketing anytime to update it."The Changelog records how the document changed. Learnings records what the team found out by shipping: which angles won, which offers flopped, which decisions were made and why. Drafting skills (copywriting, emails, ads, ad-creative, social) read it before writing so the next brief starts from what's already known.
Entry format: one row per learning.
active, watch (one data point, worth retesting), or retired.What earns a place:
What doesn't: single-run anecdotes (log those as watch or leave them out), task notes, and anything that belongs in another section. A new objection goes in Objections; a new customer phrase goes in Customer Language.
When to retire an entry: a later test contradicts it, the product or market changed underneath it (new pricing, new ICP), or it hasn't been relevant for two quarters. Change the status to retired and add a line saying why, rather than deleting it, so nobody re-tests a settled question. Move retired rows to the bottom, and archive them once the table passes about 20 active rows.
Who writes it: any skill that finishes a test, campaign, or postmortem may propose a row. Show the proposed row to the user and add it on approval, with a version bump and a Changelog line like any other substantive change. Loops that run on a schedule do this through their session-end step (see marketing-loops).
还没有评价。欢迎分享你的使用感受。
无需登录即可发表。欢迎分享使用感受、提问,或说明在哪些环境下无法运行。
还没有评论。来写第一条吧。
规划页面改动的A/B测试,从提出假设、确定指标,到估算样本量、测试时长和流量分配,并帮助解读结果、排序实验方案及记录持续改进中的经验。
为不同广告平台撰写标题、正文及静态或视频创意方案。结合投放表现和顾客反馈拓展文案,检查格式限制,并制定可衡量的测试假设,用于下一轮优化。
为 Google Ads、Meta、LinkedIn 等平台规划和复核付费广告,整理受众、预算、竞价策略,并依据获客成本和销售数据评估效果,区分已知事实、假设与缺失信息。
调查品牌在AI搜索回答中的展示和引用情况,检查内容结构、依据与网站访问设置,整理改进方向,帮助AI系统获取关键信息。
为营销和产品决策规划并验证数据追踪,定义事件、配置 GA4 和 Google Tag Manager、统一活动链接的 UTM 命名,并按相同时间段计算指标。
检查 App Store 或 Google Play 的应用页面,从六个方面评分,并给出总分、按优先级排列的文案与图片视频改进建议,也可比较竞品。