Token导航 LogoToken导航TokenDH.com
前端设计只读github未标认证来源可访问许可证需确认审计通过

charted-design图表设计

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

490

周安装

20

GitHub Stars

8

下载量

157
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:charted-design(图表设计)
来源仓库:https://github.com/marmicode/skills
仓库路径:skills/charted-design
安装命令:
npx skills add https://github.com/marmicode/skills --skill charted-design
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/marmicode/skills --skill charted-design

简介

charted-design 通过结构化访谈产出界面设计方案,涵盖功能流程、交互细节与视觉规范。

  • 适用于产品功能原型设计,输出为带编号的设计文档供团队协作评审。
  • 每次对话仅深入一个模块,需主动引导用户补充缺失信息以确保完整性。
  • 不涉及具体代码实现,重点在于需求澄清与体验路径梳理。
  • 适用宿主包括 Codex、Claude、Cursor、Gemini CLI,接入前应确认版本、权限和运行环境要求。

SKILL.md

Collaborative Design Document

Goal

Interview the user section by section to produce a design document. Use AskUserQuestion tool if available. IMPORTANT: Write the current state to design-docs/ after each section. Adapt the interview to feature complexity.

Output File

  • Path: design-docs/
  • Name: NNN-short-kebab-title.mdNNN is zero-padded (e.g. 001, 002)
  • Sequence: List files in design-docs/, take highest number, increment. Empty folder → start at 001
  • Title: Derive short kebab-case title from the feature description (ask user first)

Interview Process

For each section:

  1. Explain the section and what information is needed
  2. Ask targeted questions to gather it
  3. Summarize and propose draft content
  4. Ask user to confirm or correct before moving on

If the user answers something that belongs to a later section, acknowledge it and say you'll add it when you reach that section.

Section Order and Prompts

1. Goals

Ask: *What problem are we solving? Why does this feature matter?*

2. Non-Goals

Ask: *What is explicitly out of scope? What might people assume is included but is not?*

3. Desired Behavior

Ask: *Describe the user-visible behavior. What does the user see, click, or experience? Walk through the scenarios step by step.*

Format as a bullet list of concrete, observable behaviors.

4. Design

Ask: *How should this be implemented at a high level? What components, services, or data structures are involved?*

4a. Diagram

Produce a Mermaid flowchart of key components and interactions.

Legend:

  • Square corners = Angular components
  • Round corners = Angular services
  • Arrows: methodName({param1: Type1}): ReturnType
  • [input1: Type1] = Angular inputs
  • (output1: Type1) = Angular outputs
  • Use <br> in labels to avoid truncation

Show diagram to user and ask for corrections.

4b. Implementation Details

Ask: *Any algorithms, edge cases, or conventions?* Leave empty if nothing to add.

5. Testing Strategy

Ask: *For each component or unit from the design, what behaviors are important to test?*

Format: grouped by component/unit, each scenario with a descriptive name and arrange/act/assert steps.

Example:

## Cart component

### Displays cart items:

- Arrange fake cart repository to return 3 items: keyboard, mouse, monitor.
- Mount `Cart` component.
- Assert 3 items displayed with labels: "Keyboard", "Mouse", "Monitor".

6. PR Plan

Propose ordered, incremental PRs that:

  • Never break existing behavior
  • Are independently reviewable and mergeable
  • Keep diffs focused

Rules:

  • Scaffolding PR: If many new files, put WIP scaffolding in its own PR
  • Pre-tidy-up PR: If interfaces must change, do backward-compatible changes first (optional params, deprecations)
  • Feature PRs: Each adds one slice of user-visible or testable functionality

Include a Mermaid flowchart of PR dependencies. List each PR with a short description. Ask for feedback.

7. Alternatives Considered

Ask: *Did we consider other approaches? Why were they rejected?*

8. Kitchen Sink

Ask: *Anything else — open questions, risks, future ideas?* Leave empty if nothing.

Final Step

Assemble the full doc from the template below, write to the output file, show the user the path.

Template

# Goals

{goals}

# Non-Goals

{non_goals}

# Desired Behavior

{desired_behavior}

# Design

{design}

## Diagram

{diagram}


## Implementation Details

{implementation_details}

# Testing Strategy

{testing_strategy}

# PR Plan

{pr_dependency_diagram}


{pr_details}

# Alternatives Considered

{alternatives}

# Kitchen Sink

{kitchen_sink}

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

38.15%
按下载量换算60

Claude

27.22%
按下载量换算43

Cursor

19.33%
按下载量换算30

Gemini CLI

8.28%
按下载量换算13

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