跳到主内容
智客 ZICQ

技能库 智客分类:Agent 工作流 review-audit

Review Audit

分析正面的客户评论,以露出深层的客户见解来进行广告复制. 当用户提供客户评论并想更好地了解客户,取出VOC(客户之声),找到已准备好的语言,或用真实的客户语言构建消息策略时使用此功能. 触发任何涉及"分析这些评论","顾客在说什么","在这些评论中找到洞察力","VOC分析"等请求,或者任何想通过我的客户评论来进行创造性战略投入的变相. 输出总是由出品(如果多出品)来组织,表面有五桶洞察力:疼痛点,触发瞬间,反对,变相,和站立语言.

431 安装量

官方网址:skills.sh

技能介绍

先看中文介绍;官方 description 原文单独保留,不改写 SKILL.md。

做什么

分析正面的客户评论,以露出深层的客户见解来进行广告复制. 当用户提供客户评论并想更好地了解客户,取出VOC(客户之声),找到已准备好的语言,或用真实的客户语言构建消息策略时使用此功能. 触发任何涉及"分析这些评论","顾客在说什么","在这些评论中找到洞察力","VOC分析"等请求,或者任何想通过我的客户评论来进行创造性战略投入的变相. 输出总是由出品(如果多出品)来组织,表面有五桶洞察力:疼痛点,触发瞬间,反对,变相,和站立语言.

何时用

官方 description 未单独写出 Use when。按规范,代理会在用户任务与这段 description 的关键词匹配时激活本技能。

代理如何加载

按 Agent Skills 渐进披露:启动时只加载 name 与 description(约 100 token);任务匹配后才读入整份 SKILL.md 正文;scripts/、references/、assets/ 仅在需要时再读。 本文件正文结构:Review Audit、What You Need Before Starting、Step 1: Group by Product、Step 2: Score Review Quality (1–5)、Step 3: Filter、Step 4: Extract Insights by Bucket。 其中含规范建议的小节:分步指令。

文件分析

文件分析:这是一份仅含 SKILL.md 的指令型技能,代理激活后整份正文进入上下文。

官方 description(原文)

Analyzes positive customer reviews to surface deep customer insights for ad copy. Use this whenever a user provides customer reviews and wants to understand their customers better, extract VOC (voice of customer), find ad-ready language, or build messaging strategy from real customer language. Trigger for any request involving "analyze these reviews," "what are customers saying," "find insights in these reviews," "VOC analysis," or any variation of wanting to mine customer reviews for creative strategy inputs. Output is always organized by product (if multiple), and surfaces five buckets of insight: pain points, trigger moments, objections, transformations, and standout language.

Review AuditWhat You Need Before StartingStep 1: Group by ProductStep 2: Score Review Quality (1–5)Step 3: FilterStep 4: Extract Insights by BucketBucket 1: Pain PointsBucket 2: Trigger MomentsBucket 3: Objections Before PurchasingBucket 4: TransformationsBucket 5: Standout Language & Ad-Ready PhrasesOutput Format

来源分类:skills.sh agent-skill

SKILL.md 与 Agent 调用

官方规范 ↗
name
review-audit
description
Analyzes positive customer reviews to surface deep customer insights for ad copy. Use this whenever a user provides customer reviews and wants to understand their customers better, extract VOC (voice of customer), find ad-ready language, or build messaging strategy from real customer language. Trigger for any request involving "analyze these reviews," "what are customers saying," "find insights in these reviews," "VOC analysis," or any variation of wanting to mine customer reviews for creative strategy inputs. Output is always organized by product (if multiple), and surfaces five buckets of insight: pain points, trigger moments, objections, transformations, and standout language.
  1. 发现技能客户端向 Agent 提供名称与描述目录。
  2. 匹配与调用用户指定或任务匹配后,载入 SKILL.md 指令。
  3. 按需加载按步骤读取参考文档、使用脚本与素材。

具体调用语法与可用工具以目标 Agent 客户端为准。 查看调用机制说明 ↗

安装这个技能

Skills CLI ↗

先选择目标 Agent 和安装范围,保留技能包的附属文件,安装后检查客户端能否发现该技能。

交给 Agent 安装

复制安装指令给支持 Agent Skills 的代理,确认其中的目标目录与客户端匹配。

把 Agent Skill「review-audit」安装到我的项目:SKILL.md 原文与官方 description 见 https://zicq.com/zh/skills/skl-fca66d928d0642b6-%E5%AE%A1%E6%9F%A5-%E5%AE%A1%E8%AE%A1.html
请存为 .cursor/skills/review-audit/SKILL.md 或 .claude/skills/review-audit/SKILL.md,frontmatter 的 name 与 description 保持原样,不要改写。

GitHub 完整包 ↗

终端安装 · Skills CLI

需要 Node.js 与 npx。先查看仓库技能列表,确认实际名称。

npx skills add 'https://github.com/motion-team/creative-strategy-skills' --list

npx skills add 'https://github.com/motion-team/creative-strategy-skills' --skill 'review-audit'

CLI 会交互选择目标 Agent,默认安装到项目;用户级安装使用 -g。先通过查看命令核对仓库内容,再用 npx skills list 检查已安装技能。

阅读排版

name: review-audit description: 'Analyzes positive customer reviews to surface deep customer insights for ad copy. Use this whenever a user provides customer reviews and wants to understand their customers better, extract VOC (voice of customer), find ad-ready language, or build messaging strategy from real customer language. Trigger for any request involving "analyze these reviews," "what are customers saying," "find insights in these reviews," "VOC analysis," or any variation of wanting to mine customer reviews for creative strategy inputs. Output is always organized by product (if multiple), and surfaces five buckets of insight: pain points, trigger moments, objections, transformations, and standout language.'

Review Audit

This system mines positive customer reviews to extract the insights that make ad copy actually work — the real language, real pain, real moments, and real transformations that customers experienced. The output feeds directly into creative strategy and hook writing.

The goal is not to summarize reviews. The goal is to find the raw material for ads.


What You Need Before Starting

Reviews can be provided in any format:

  • Pasted directly into the chat
  • CSV or spreadsheet upload
  • Copy/pasted from a document

If multiple products are present in the review set, identify them before beginning. All output is separated by product.

If the format is unclear or product attribution is ambiguous, ask before proceeding.


Step 1: Group by Product

If the brand sells multiple products, sort all reviews by product first. Every subsequent step runs separately per product.

If all reviews are for a single product, skip grouping and proceed.


Step 2: Score Review Quality (1–5)

Before analysis, score every review for quality. This determines what gets analyzed and what gets discarded.

| Score | What it looks like | |---|---| | 1 | Garbage — gibberish, swear words, 2–3 meaningless words, zero signal ("great product", "love it", "👍") | | 2 | Low signal — very short, vague, no specific detail or emotion | | 3 | Moderate — mentions the product, some specificity, but no vivid detail or emotional depth | | 4 | High quality — specific, describes a real experience, references a before/after or a feeling | | 5 | Gold — long, emotional, vivid, paragraph-level detail; the customer was so moved they wrote an essay about it |

Score 5 reviews are the priority. They contain the most usable language and the deepest insight.


Step 3: Filter

Discard all reviews scored 1. Do not include them in analysis.

Analyze scores 2–5, with emphasis on 4s and 5s. Low-scoring reviews (2–3) can contribute to pattern identification but should not be the source of pulled quotes.


Step 4: Extract Insights by Bucket

Run this analysis separately for each product. Within each bucket, group similar insights together and write a brief summary of the pattern. Then pull the best word-for-word quotes that exemplify it.

Do not editorialize the quotes. Pull them exactly as written.


Bucket 1: Pain Points

What problem were they experiencing before they found this product?

Look for: descriptions of the problem they had, how long they'd had it, what they'd tried before, how it affected their life, the emotional weight of living with it.

For each pain theme identified:

  • Name the theme
  • Write a 2–3 sentence summary of the pattern across reviews
  • Flag the strongest quotes for the swipe file

Bucket 2: Trigger Moments

What finally made them buy?

Look for: the specific moment, event, or realization that pushed them over the edge. This is the thing that turned a maybe into an add-to-cart. It could be a life event (wedding, diagnosis, vacation), a recommendation (friend, doctor, TikTok), hitting a breaking point, or running out of patience with other solutions.

For each trigger theme identified:

  • Name the theme
  • Write a 2–3 sentence summary of the pattern
  • Flag the strongest quotes for the swipe file

Bucket 3: Objections Before Purchasing

What almost stopped them from buying?

Look for: skepticism they mention having had, comparisons to other products they'd tried, price hesitation, disbelief that this would actually work, fear of wasting money again.

Note: In positive reviews, objections are almost always mentioned in past tense — "I was skeptical but..." or "I almost didn't try it because..." These are gold for objection-handling ad copy.

For each objection theme identified:

  • Name the theme
  • Write a 2–3 sentence summary of the pattern
  • Flag the strongest quotes for the swipe file

Bucket 4: Transformations

What changed for them after using the product?

Look for: the specific result they experienced, how their life is different now, the emotional shift (confidence, relief, freedom, pride), and — most importantly — how they describe the transformation in their own words. The more specific and visceral, the better.

For each transformation theme identified:

  • Name the theme
  • Write a 2–3 sentence summary of the pattern
  • Flag the strongest quotes for the swipe file

Bucket 5: Standout Language & Ad-Ready Phrases

Exact language worth stealing for ads.

This bucket is different from the others. It is not organized by theme — it is a curated collection of the most vivid, emotionally charged, specific, and scroll-stopping phrases pulled from across all buckets. These are the lines that made you stop while reading. The ones that don't need to be rewritten. The ones a copywriter would highlight and build an ad around.

Pull these verbatim. Note which product they're from.

What to look for:

  • Unusually specific descriptions of pain or transformation
  • Phrases that capture an emotion in a way you couldn't have written yourself
  • Before/after language that is visceral and concrete
  • Lines that could work as a hook with zero editing
  • Anything that made you feel something while reading it

Output Format

Produce a separate full output for each product. Structure:

─────────────────────────────────────
PRODUCT: [Product Name]
Reviews analyzed: [X] | Discarded (score 1): [X]
─────────────────────────────────────

BUCKET 1: PAIN POINTS

[Theme Name]
Summary: [2–3 sentences on the pattern]

[Theme Name]
Summary: [2–3 sentences on the pattern]

---

BUCKET 2: TRIGGER MOMENTS

[Theme Name]
Summary: [2–3 sentences on the pattern]

---

BUCKET 3: OBJECTIONS BEFORE PURCHASING

[Theme Name]
Summary: [2–3 sentences on the pattern]

---

BUCKET 4: TRANSFORMATIONS

[Theme Name]
Summary: [2–3 sentences on the pattern]

---

BUCKET 5: STANDOUT LANGUAGE & AD-READY PHRASES

"[Exact quote]"
"[Exact quote]"
"[Exact quote]"
[etc.]

─────────────────────────────────────

All word-for-word quotes are collected in Bucket 5. Do not scatter quotes throughout buckets 1–4 — keep the summaries clean and let the swipe file be the dedicated place for raw language.


How This Feeds the Rest of the Stack

The output of this analysis plugs directly into creative strategy and execution:

  • Pain Points → Creative Strategy Engine — pain buckets map directly to the pain/desire anchor layer
  • Trigger Moments → Hook Writing — trigger moments are often the most powerful hook material; they capture the exact moment of emotional readiness
  • Objections → Hook Writing / Creative Mechanics — objections inform Borrowed Enemy, Reframe, and Risk Reversal mechanics and hook tactics
  • Transformations → Hook Writing — transformation language feeds aspirational and social proof hooks
  • Standout Language → Hook Voice Patterns — the best phrases can be added directly to the swipe file as native voice patterns pulled from real customers

Notes on Quality

  • Score 5 reviews should be read in full and treated as primary sources
  • Score 2–3 reviews are useful for pattern confirmation but not quote sourcing
  • If the review set is small (under 20 reviews), note this — patterns may not be statistically meaningful but language is still usable
  • If a product has too few quality reviews to surface meaningful patterns, flag it rather than manufacturing themes that aren't there

相关技能

Agent 工作流

Skill Creator

创造有效技能指南。 当用户想创造出新的技能(或更新现有的技能),以专业知识,工作流程,或工具集成来扩展克洛德的能力时,应该使用这种技能.

Agent 工作流

Clawdhub

使用ClawdHub CLI搜索,安装,更新并发布从taladhub.com的代理技能. 需要获取苍蝇上的新技能时使用,将安装的技能同步到最新版本或特定版本,或者发布 npm-instainddhub CLI 的新/更新的技能文件夹.

Agent 工作流

Agent Team Orchestration

管弦乐团多代理团队,任务设定周期,交接协议,审查工作流程. 使用时间: (1)建立2+特派员队伍,具有不同专业,(2)确定任务路线和生命周期(收录框_ spec_建设_审查_完成),(3)在特派员之间制定交接协议,(4)建立审查和质量关口,(5)管理特派员之间的交流和文物共享.

Agent 工作流

Superpowers

Spec-first,TDD,子代理驱动的软件开发工作流程. 当:(1)构建任何新功能或应用——触发脑暴_计划_子代理执行回路,(2)调试出一个bug或测试失败——触发系统性的根起过程,(3)用户说"让我们构建","帮助我计划","我想添加X",或"这个被打破",(4)完成一个功…