Token导航 LogoToken导航TokenDH.com
研究检索只读github未标认证来源可访问许可证需确认审计通过

growth-design-review增长设计审查

Agent Skill

用于辅助界面设计、视觉规范、排版、配色、布局和交互体验优化。它适合让 Agent 根据产品场景整理页面结构、生成 UI 方案、检查视觉一致性或改进组件层级。使用时需要结合现有品牌、设计系统和用户任务,不应只堆装饰元素;涉及真实页面改动时,应通过截图或浏览器预览检查文本溢出、对齐和响应式表现。

总安装

783

周安装

32

GitHub Stars

11

下载量

253
CodexClaudeCursorGemini CLI

安装说明

本站只整理中文说明和来源信息,不托管安装包,也不代用户安装。

GitHub

来源数

2

许可证

unknown

最后核验

2026-05-01

来源状态

来源可访问

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:growth-design-review(增长设计审查)
来源仓库:https://github.com/woakin/growth-design-review
仓库路径:skills/growth-design-review
安装命令:
npx skills add https://github.com/woakin/growth-design-review --skill growth-design-review
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/woakin/growth-design-review --skill growth-design-review

简介

该技能辅助界面设计、视觉规范和交互体验优化,适合生成 UI 方案或检查一致性。

  • 适用于页面结构整理、组件层级改进和响应式表现评估。
  • 使用时需结合品牌系统和用户任务,避免堆砌装饰元素。
  • 涉及真实页面改动时应通过截图或预览检查文本溢出和对齐问题。
  • growth-design-review 属于研究检索类 Skill,可作为该场景下的辅助能力补充。

SKILL.md

🧠 Growth Design Review Skill

Review UI designs or code against Product Psychology & Growth Design frameworks. Propose a prioritized development plan following each framework's best practices to maximize the Net Perceived Value.


📚 Frameworks Overview

This skill applies five core frameworks plus an AI extension. CRITICAL: To optimize context usage, only load the framework definitions you need for the task at hand from the reference/ directory.

  1. Psych Framework — Maximize Net Perceived Value (Motivation - Friction).
  2. B.I.A.S. Framework — Behavioral audit: Block, Interpret, Act, Store.
  3. C.L.E.A.R. Scorecard — Copywriting, Layout, Emphasis, Accessibility, Reward.
  4. UI Rules of Thumb — Page-type specific patterns (e.g., Landing, Dashboard, Onboarding).
  5. Psychological Triggers — Utilization of the IKEA Effect, Zeigarnik Effect, and Loss Aversion.

🛠 Review Process

Whenever a user requests a review, strictly follow these steps:

  1. Diagnostic Routing: Determine if the interface involves AI (LLMs, generative UI, or autonomous agents).

- *If standard SaaS/Web:* Apply the Psych Framework, B.I.A.S. Audit, and C.L.E.A.R. Scorecard. - *If AI-Native:* Apply standard frameworks to diagnose friction, but you MUST propose solutions using specific Shape of AI patterns.

  1. Context Routing (Lazy Load): Analyze the user's prompt to determine which frameworks are needed. Use file reading tools to load ONLY the specific markdown files required from this skill's directory:

- If user asks for Psych -> read reference/psych.md - If user asks for B.I.A.S. -> read reference/bias.md - If user asks for C.L.E.A.R. -> read reference/clear.md - If user asks for Rules of Thumb -> read reference/rules-of-thumb.md - If user asks for Psychological Triggers -> read reference/psychological-triggers.md - If user asks for a "Full Audit" -> Read all core framework files individually. - If the interface is AI-Native -> read reference/shape-of-ai.md

  1. Understand the Target: Read the target UI files (components, pages, styles) the user wants reviewed. If none are provided, ask the user which files to evaluate.
  2. Systematic Audit: Evaluate the UI against *all* applicable frameworks systematically.
  3. Scoring: Rate each C.L.E.A.R. dimension (Copy, Layout, Emphasis, Accessibility, Reward) on a strict 1–5 scale.
  4. Actionable Output: Generate the output exactly matching the format below.
  5. Next Steps: Offer to implement the Priority 1 (Critical) items immediately for the user.

📋 Required Output Format

You must format your response exactly as follows:

## 📈 Growth Design Audit

### 1. Psych Framework Analysis
**Net Perceived Value Assessment:**
- **Motivation signals found:** [List any]
- **Friction points found:** [List any]
- **Psych Additions present:** [Rewards, progress markers, delighters]
- **Psych Subtractions to fix:** [Redundant fields, cognitive load, jargon]
- **Labor Illusion usage:** [Present/missing, recommendation]

### 2. B.I.A.S. Behavioral Audit
| Dimension | Status | Finding |
| :--- | :---: | :--- |
| **Block** (Visual Hierarchy & Trust) | [Pass/Fail] | [Brief explanation] |
| **Interpret** (Clarity over Cleverness) | [Pass/Fail] | [Brief explanation] |
| **Act** (Decision Simplicity) | [Pass/Fail] | [Brief explanation] |
| **Store** (Peak-End Rule) | [Pass/Fail] | [Brief explanation] |

### 3. C.L.E.A.R. Scorecard
| Dimension | Score (1-5) | Finding |
| :--- | :---: | :--- |
| **C** - Copywriting | X/5 | [Brief explanation] |
| **L** - Layout | X/5 | [Brief explanation] |
| **E** - Emphasis | X/5 | [Brief explanation] |
| **A** - Accessibility | X/5 | [Brief explanation] |
| **R** - Reward | X/5 | [Brief explanation] |
| **Total Score** | **XX/25** | |

### 4. UI Rules of Thumb
- **Page Type Identified:** [Landing / Dashboard / Onboarding / Component / Other]
- **Compliance Analysis:** [Findings per applicable rule based on the page type]

### 5. Psychological Triggers
- **IKEA Effect:** [Present / Missed Opportunity] - [Explanation]
- **Zeigarnik Effect:** [Present / Missed Opportunity] - [Explanation]
- **Loss Aversion:** [Present / Missed Opportunity] - [Explanation]

### 6. Shape of AI Analysis (If AI-Native)
| Dimension | Friction Point | NPV Impact | Recommended Shape of AI Pattern |
| :--- | :--- | :--- | :--- |
| [Category] | [Brief Description] | [Impact] | **[Pattern Name]:** [Execution suggestion] |

---

## 🏗 Development Plan

### 🔴 Priority 1 — Critical (High Impact, Low Effort)
| # | Action | Framework | File(s) | Rationale |
|---|--------|-----------|---------|-----------|
| 1 | [Action] | [Framework] | [File path] | [Why this matters] |

### 🟡 Priority 2 — Important (High Impact, Higher Effort)
| # | Action | Framework | File(s) | Rationale |
|---|--------|-----------|---------|-----------|
| 1 | [Action] | [Framework] | [File path] | [Why this matters] |

### 🟢 Priority 3 — Nice to Have (Lower Impact)
| # | Action | Framework | File(s) | Rationale |
|---|--------|-----------|---------|-----------|
| 1 | [Action] | [Framework] | [File path] | [Why this matters] |

适合场景

01

用户想查找某类 Agent Skill 时

02

需要根据任务场景推荐可安装能力包时

03

需要对比不同来源的安装命令和来源信息时

能力概览

能力 1

按任务关键词查找相关 Skills

能力 2

展示可复制的安装命令

能力 3

保留来源站点、仓库和原始说明,方便继续核验

能力 4

展示第三方安全扫描或审计结果

安装后应在对应宿主中按原始 README 的触发条件使用;具体调用方式请以来源页面和 README 为准。

平台分布

Codex

33.5%
按下载量换算85

Claude

30.03%
按下载量换算76

Cursor

19.4%
按下载量换算49

Gemini CLI

9.02%
按下载量换算23

安全审计

Gen Agent Trust Hub

通过

Socket

通过

Snyk

通过

权限和风险

只读

该 Skill 主要提供规则、说明或参考内容,本身偏只读;真正读写文件、联网或执行命令仍取决于宿主 Agent 的任务。

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。当前只有一个来源,正式发布前建议补源仓库或其他目录站核验。

来源信息

继续浏览同类 Skills