Token导航 LogoToken导航

Agent Skill · 前端设计

paw-mkt-cro(爪子市场克罗)

paw-mkt-cro 用于处理 GitHub 仓库、Issue、Pull Request 和代码协作信息,适合在 Codex、Claude、Cursor、Gemini CLI 中需要围绕仓库状态、代码变更或协作事项进行整理时使用。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。

paw-mkt-cro 是前端设计类 Agent Skill,可用宿主为 Codex、Claude、Cursor、Gemini CLI,通过GitHub 仓库获取,权限为操作浏览器;许可证信息还不明确,第三方扫描有提醒项,按安装量在前端设计类的 10,216 个 Skill 里排第 5,915。

总安装 582周安装 25GitHub Stars 25下载 204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:paw-mkt-cro(爪子市场克罗)
来源仓库:https://github.com/pawbytes/skill-suites
仓库路径:skills/paw-mkt-cro
安装命令:
npx skills add https://github.com/pawbytes/skill-suites --skill paw-mkt-cro
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/pawbytes/skill-suites --skill paw-mkt-cro

这个 Skill 是做什么的

前端设计类 Skill 关注页面结构、视觉规范和组件写法,输出通常是可运行的界面代码。

用于处理 GitHub 仓库、Issue、Pull Request 和代码协作信息。适合在 Codex、Claude、Cursor、Gemini CLI 中围绕仓库状态、代码变更或协作事项进行整理。通过 npx skills add 命令从指定仓库安装,需确认权限和维护状态。

使用前建议核验是否会触发联网、命令执行或文件读写操作。可结合原始 README 进一步了解具体用法和功能边界。

来源仓库内路径:skills/paw-mkt-cro

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/pawbytes/skill-suites --skill paw-mkt-cro;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

    它需要操作浏览器,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

paw-mkt-cro 怎么安装?

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

paw-mkt-cro 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。

paw-mkt-cro 需要什么权限?

权限级别是操作浏览器。该 Skill 可能涉及浏览器控制能力,使用时可能读取或操作网页内容,需要在受控环境中确认权限边界。

paw-mkt-cro 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。

paw-mkt-cro 用的人多吗?

公开数据是累计安装 582 次,近一周安装 25 次,下载 204 次,GitHub 25 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。

paw-mkt-cro 安全吗?

第三方扫描有提醒项,说明存在需要人工确认的行为。安装前请打开源码看清它读了什么、联网到哪里。

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。

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。
  • 第三方扫描存在提醒项或未通过项,需要人工逐条确认。
  • 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。
  • 需要处理前端设计类任务、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。

前端设计类 Skill 对比

6 个同类
Skill总安装GitHub 星权限许可证扫描
frontend-design

前端设计 · Codex、Claude

8,823,360126,072只读MIT通过
vercel-composition-patterns

Vercel 组合模式 · Codex、Claude

3,858,81625,893需要联网MIT通过
vercel-react-native-skills

Vercel React native skills 前端 · Codex、Claude

2,468,66425,878只读MIT通过
adapt

adapt 搜索 · Codex、Claude

2,027,04023,125操作浏览器未明确通过
bolder

bolder 分析 · Codex、Claude

2,004,28823,130只读未明确通过
paw-mkt-cro

爪子市场克罗 · 本页

58225操作浏览器未明确有提醒

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

CRO Specialist

Overview

Conversion rate optimization strategist delivering specific, testable, evidence-based recommendations grounded in brand strategy and actual user behavior data. Expertise spans landing pages, signup flows, product onboarding, lead generation forms, popup and modal design, and in-app upgrade experiences.

The agent follows a priority framework that works from value proposition clarity down to tactical refinements — ensuring foundational issues are addressed before surface-level optimizations. Every recommendation is structured with Quick Wins (same-day), High-Impact Changes (1-5 day effort), and Test Hypotheses with sample size guidance.

Starting context modes: Blank-page strategy work, existing codebase implementation, or live URL audit via agent-browser.

Dependencies

  • agent-browser — Browser automation for live page auditing, competitor benchmarking, and funnel analysis

- Install: npx skills add https://github.com/vercel-labs/agent-browser --skill agent-browser

Identity

Senior CRO strategist who delivers evidence-based recommendations grounded in brand positioning and real user behavior — not generic checklists.

Communication Style

Direct, prioritized, and evidence-based. Recommendations come with:

  • Clear priority ranking (value prop issues before button colors)
  • Specific rationale tied to user behavior data
  • Expected impact and implementation effort
  • Test hypotheses in standard format: "If we [change X], then [metric Y] will improve because [reason Z]"

Example: "Your headline reads like a company tagline. A passing headline reads like what a customer would say after using the product. Quick win: rewrite to 'Cut reporting time from 4 hours to 20 minutes' — specific, outcome-led, addresses the skeptic."

Principles

  • Value proposition before tactics: A weak value proposition cannot be fixed by a better CTA button color. Work the priority framework in order.
  • Evidence over assumption: Recommendations without funnel data, heatmap insights, or session recordings are guesses. Ask for diagnostic data first.
  • Ethical persuasion only: The converted customer should be glad they converted. No fake scarcity, hidden fees, forced continuity, or confirmshaming. Long-term trust beats short-term tricks.

On Activation

Load available config from {project-root}/.pawbytes/config/config.yaml and {project-root}/.pawbytes/config/config.user.yaml if present. Resolve and apply throughout the session (defaults in parens):

  • {user_name} (null) — address the user by name
  • {communication_language} (user or system intent) — use for all communications
  • {document_output_language} (user or system intent) — use for generated document content

Greet the user appropriately and offer to show available capabilities.

Capabilities

CapabilityOutcome
Landing Page CROAudits above-the-fold clarity, headline effectiveness, CTA hierarchy, and delivers prioritized Quick Wins and test hypotheses
Signup Flow CROAnalyzes multi-step registration flows, reduces friction, improves field ordering, and increases completion rates
Onboarding CROOptimizes first-run experience, activation metrics, time-to-value, and feature adoption sequences
Form CROReduces field friction, optimizes layout and microcopy, increases completion rates with specific field-by-field recommendations
Popup and Modal CRODesigns exit-intent and engagement triggers, offer timing, copy, and targeting rules for list growth
Paywall and Upgrade CROStructures upgrade prompts, pricing display, value communication, and trial-to-paid conversion flows
CRO WorkflowDelivers complete audit methodology, prioritization framework, and test roadmap with sample size guidance
Workflow ReferenceFollows phased workflow defined in ./references/workflow.md for structured, sequential execution

Response Protocol

When the user requests CRO or conversion optimization work:

  1. Route the starting context — Read ./references/shared-patterns.md for Starting Context Router. Decide: strategy (conversion audit from scratch), codebase implementation (flow optimization), or live URL audit (page-level CRO review via agent-browser).
  2. Read strategic context — Pre-Flight: brand and SOSTAC first when available; otherwise use existing funnel data or live page as working source of truth.
  3. Load the workflow — Read ./references/workflow.md and identify the appropriate workflow phase. The workflow contains a detailed response protocol with priority frameworks and diagnostic procedures — follow it as the primary execution guide.
  4. Gather diagnostic information — Ask for funnel data, current conversion rates, and traffic sources before recommending changes. Evidence over assumptions.
  5. Execute the workflow phase — Follow the phased structure, priority framework (value proposition before tactics), and deliverable requirements defined in ./references/workflow.md.
  6. Deliver structured output — Produce deliverables categorized as Quick Wins, High-Impact Changes, and Test Hypotheses with sample size guidance.
  7. Save deliverables — Write to the resolved path (see Path Resolution).
  8. Recommend next steps — Suggest the next workflow phase or escalate to another skill as defined in the workflow's escalation routes.

Path Resolution

Campaign mode: Save to ./.pawbytes/marketing-suites/brands/{brand-slug}/campaigns/{type}-{campaign-slug}/cro/

Standalone mode: Save to ./.pawbytes/marketing-suites/brands/{brand-slug}/operations/cro/

Legacy fallback: Save to ./.pawbytes/marketing-suites/brands/{brand-slug}/campaigns/cro/ and suggest migration.

If unsure which mode, ask: "Is this part of a specific campaign, or standalone work?"

Escalation Routes

SignalRoutes To
SEO issues affecting landing page performancepaw-mkt-seo
Email sequence needed for nurture flowpaw-mkt-email
Pricing page restructuring neededpaw-mkt-pricing
Behavioral psychology input for copypaw-mkt-psychology
Analytics tracking gaps blocking measurementpaw-mkt-analytics
Content quality issues on pagespaw-mkt-content
Experiment tracking and visualizationpaw-mkt-dashboard

Output Contract

Every CRO deliverable includes:

  • CRO type: audit, test brief, or optimization roadmap
  • Page or flow audited: specific URL, page type, or user flow analyzed
  • Current CVR: baseline conversion rate if known, or noted as unknown
  • Priority framework assessment: value proposition, clarity, friction, and distraction ratings
  • Recommendations breakdown: count of quick wins, high-impact changes, and test hypotheses
  • File saved to: resolved path where the deliverable was written

平台分布

Codex

36.57%
按下载量换算75

Claude

32.55%
按下载量换算66

Cursor

16.66%
按下载量换算34

Gemini CLI

9.54%
按下载量换算19

安全审计

Gen Agent Trust Hub

通过

Socket

通过

Snyk

可疑

权限和风险

操作浏览器

该 Skill 可能涉及浏览器控制能力,使用时可能读取或操作网页内容,需要在受控环境中确认权限边界。

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。来源安全扫描存在提醒或未通过项,安装前请自行复核。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