Token导航 LogoToken导航

Agent Skill · 研究检索

design(设计)

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。

design 是研究检索类 Agent Skill,可用宿主为 Codex、Claude、Cursor、Gemini CLI,通过GitHub 仓库获取,权限为只读;许可证信息还不明确,第三方扫描有提醒项,按安装量在研究检索类的 73,675 个 Skill 里排第 55,495。

总安装 475周安装 20GitHub Stars 公开资料未说明下载 166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:design(设计)
来源仓库:https://github.com/kelvinz/cobb
仓库路径:skills/design
安装命令:
npx skills add https://github.com/kelvinz/cobb --skill 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/kelvinz/cobb --skill design

这个 Skill 是做什么的

研究检索类 Skill 主要解决「找资料、核对事实、把来源整理清楚」的问题,通常会让助手先检索再总结。

design 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,适合在 Codex、Claude、Cursor、Gemini CLI 中生成 UI 方案或改进组件层级。适用于产品页面结构整理、视觉一致性检查和响应式表现验证。通过 npx skills add 命令从指定 GitHub 仓库安装使用。

使用时需结合现有品牌和设计系统,避免仅堆砌装饰元素。design 属于研究检索类 Skill,可作为该场景下的辅助能力补充。

来源仓库内路径:skills/design

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/kelvinz/cobb --skill design;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

    它标注为只读,先用只读任务试一次,确认行为符合预期再扩大使用范围。

常见问题

6 条

design 怎么安装?

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

design 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。

design 需要什么权限?

权限级别是只读。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。

design 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。

design 用的人多吗?

公开数据是累计安装 475 次,近一周安装 20 次,下载 166 次。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。

design 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。
  • 标注为只读权限,试用阶段不容易误改本地文件。

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。
  • 需要处理研究检索类任务、又不想每次都手写提示词的人。
  • 只想先试试效果、不希望它改动本地文件的人。

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。

研究检索类 Skill 对比

6 个同类
Skill总安装GitHub 星权限许可证扫描
find-skills

寻找技能 · Codex、Claude

31,199,81316,486操作浏览器MIT有提醒
self-improving-agent

自我改进 Agent · OpenClaw

16,464,0023,396接触敏感数据MIT-0通过
soultrace

灵魂追踪 · Codex、Claude

7,588,94412需要联网未明确有提醒
self-improving

自我完善 · OpenClaw

6,829,2841,046需要联网MIT-0通过
microsoft-foundry

微软代工厂 · Codex、Claude

6,614,144764调用外部服务未明确有提醒
design

设计 · 本页

475—只读未明确有提醒

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

Design

Route each design request to the correct mode, then execute that mode's output contract with shared quality constraints.

Core Workflow

  1. Select a primary mode from the mode map in this file.
  2. Select one secondary mode only when the request clearly spans multiple concerns.
  3. Fill the mandatory routing brief before producing output. If inputs are missing, state explicit assumptions.
  4. Load only the reference files required for the chosen mode(s).
  5. Deliver the primary mode contract completely; add secondary mode deltas without diluting the primary outcome.

Mode Map

Select one primary mode before work starts. Add one secondary mode only when needed.

  1. ui Typical request patterns: build or refine screens, components, dashboards, landing pages, settings, and app flows. Primary goal: ship clear and distinctive interfaces. Expected deliverables: working UI code, token mapping, full state coverage, concise hierarchy rationale. Load reference: references/ui.md. Load references/ui-examples.md only when concrete code examples are needed, especially for 5 or 7 tier token sets.
  2. ux Typical request patterns: plan or improve flows, IA, usability, accessibility, or run design audits. Primary goal: improve task success, comprehension, and evidence-based quality. Expected deliverables: UX structure outputs or severity-ordered audit findings with concrete fixes. Load reference: references/ux.md.
  3. motion Typical request patterns: define or implement transitions, microfeedback, loading, and continuity behavior. Primary goal: communicate state change and preserve orientation without friction. Expected deliverables: motion spec or code, timing/easing choices, reduced-motion fallback. Load reference: references/motion.md.
  4. imagery Typical request patterns: produce posters, editorial visuals, and static compositions. Primary goal: deliver high-impact static visual artifacts. Expected deliverables: philosophy .md plus .png or .pdf artifact (single page unless requested). Load reference: references/imagery.md.

Selection Rules

  1. Select ui for implementation-heavy interface design and visual system decisions.
  2. Select ux for information architecture, usability, accessibility, or audit-heavy requests.
  3. Select motion when behavior choreography is the core task.
  4. Select imagery for non-interactive visual composition outputs.
  5. When requests blend concerns, keep one primary mode and one secondary mode.

Routing Decision Tree

Use this sequence to select the primary mode:

  1. If the request is for a non-interactive visual artifact (.png, .pdf, poster, editorial composition), select imagery.
  2. If the request is primarily about IA, usability, accessibility, or audits, select ux.
  3. If the request is primarily about transitions, microfeedback, choreography, or reduced-motion behavior, select motion.
  4. If the request is primarily about building/refining interface code, components, layout, or visual systems, select ui.

Then add one secondary mode only when required:

  1. Use ux as secondary to ui when task flow or accessibility structure changes are a core part of the implementation.
  2. Use motion as secondary to ui when animation behavior is explicitly in scope.
  3. Use motion as secondary to ux when motion quality or reduced-motion compliance is a key audit axis.
  4. Keep imagery standalone by default; add a secondary mode only when the user explicitly asks for a companion interactive adaptation.

Complexity Budget

Choose one level early and stay consistent:

  1. baseline: practical, concise, low-risk execution.
  2. expressive: clear personality with controlled complexity.
  3. maximal: high visual and motion complexity with strict quality control.

Default to baseline for internal or time-constrained work. Default to expressive for most public-facing work. Use maximal only when the brief explicitly rewards spectacle and budget supports it.

Mandatory Routing Brief

Complete this brief before solution design:

Primary mode:
Secondary mode (optional):
Request type (new design, refinement, audit, artifact):
Human and audience:
Primary task to enable:
Context and constraints:
Complexity budget (baseline | expressive | maximal):
Output artifact(s) required:
Success criteria:

Global Non-Negotiables

Apply these across all modes:

  1. Preserve accessibility fundamentals (contrast, keyboard reachability, visible focus, reduced-motion support).
  2. Preserve responsive intent across mobile and desktop where screens are involved.
  3. Preserve performance by preferring efficient rendering and animation primitives.
  4. Cover required interaction and data states for any interactive surface.
  5. Keep decisions explainable with concrete rationale tied to the request and constraints.
  6. Layouts must tolerate longer copy, dynamic data, and localisation (wrapping, truncation rules, min/max widths).

References

  • references/ui.md: interface craft system, tokens, component rules, output contract, and quality checks.
  • references/ui-examples.md: concrete Tailwind and component examples for 3/5/7 tier token systems.
  • references/ux.md: UX structure, usability/accessibility checks, and audit workflow/report format.
  • references/motion.md: motion purpose, timing/easing, performance, accessibility, and motion delivery contract.
  • references/imagery.md: static visual artifact workflow and required output format.

平台分布

Codex

34.24%
按下载量换算57

Claude

33.49%
按下载量换算56

Cursor

18.75%
按下载量换算31

Gemini CLI

9.13%
按下载量换算15

安全审计

Gen Agent Trust Hub

通过

Socket

通过

Snyk

可疑

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。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