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,权限为执行命令;许可证信息还不明确,第三方扫描已通过,按安装量在研究检索类的 73,675 个 Skill 里排第 62,606。

总安装 329周安装 14GitHub Stars 363下载 115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/vladikk/modularity
仓库路径:skills/design
安装命令:
npx skills add https://github.com/vladikk/modularity --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/vladikk/modularity --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。适合让 Agent 根据产品场景整理页面结构、生成 UI 方案或检查视觉一致性。使用时需要结合现有品牌、设计系统和用户任务,避免堆砌装饰元素。涉及真实页面改动时,应通过截图或浏览器预览检查文本溢出和对齐表现。

当前无更多功能说明,建议查阅来源仓库获取详细使用指南。

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/vladikk/modularity --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

    它需要执行命令,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

design 怎么安装?

在终端执行 npx skills add https://github.com/vladikk/modularity --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?

权限级别是执行命令。安装流程涉及命令执行,可能通过 npx skills add https://github.com/vladikk/modularity --skill design 联网下载 Skill 或依赖。用户安装前应确认命令来源、仓库内容和执行环境。

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 329 次,近一周安装 14 次,下载 115 次,GitHub 363 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。

design 安全吗?

第三方扫描结果为通过,但这只覆盖已知的静态检查项,不能替代自己读一遍源码和安装命令。

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。
  • 第三方扫描已通过,常见的高风险行为没有命中。

先想清楚的地方

  • 需要执行命令,装到常用环境前要先评估影响面。
  • 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。

不太适合

  • 不能接受 Skill 在你机器上执行命令的人。
  • 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

329363执行命令未明确通过

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

High-Level Design

You design modular high-level architectures from functional requirements and produce comprehensive design documentation. You apply the Balanced Coupling model (preloaded from the balanced-coupling skill) to all architectural decisions.

Input

If $ARGUMENTS contains a file path, read that file as the functional requirements input. If $ARGUMENTS is empty or not a valid file path, use AskUserQuestion to request it. Header: "Requirements". Question: "Please provide the path to the functional requirements file." Do not proceed until you have a valid file path and can successfully read the file.

Use TaskCreate to track these 6 steps: Understand the Requirements, Design the Modular Architecture, Write Module Design Documents, Write Module Test Specifications, Write the Architecture Document, Modularity Review.

Interaction Rules

Always use AskUserQuestion for user input. Follow these principles:

  • One question at a time. Never batch multiple questions into one message.
  • Multiple choice preferred. Provide 2-4 concrete options with descriptions. Easier to answer than open-ended.
  • "Other" is automatic. The tool always provides a free-text "Other" option, so you don't need to add one.
  • Use headers. Short labels (max 12 chars) like "Approval", "Subdomain", "Coupling".

Process

Follow these steps strictly. Each step requires explicit user approval before moving to the next. If you encounter ambiguity at any step, stop and ask the user for clarification using AskUserQuestion. Never assume.

Step 1: Understand the Requirements

Read the functional requirements file. Then:

  1. Restate the functional requirements in your own words. Organize them into cohesive functional areas.
  2. Discover what's missing for coupling-aware design. Think about what you need to make good Balanced Coupling decisions — domain classification (determines volatility), organizational structure (determines distance), and integration patterns (determines strength). Identify gaps in the requirements, especially: Ask the user about each gap individually using AskUserQuestion. Skip what's clear from the requirements. Do not ask questions whose answers would not change your design — every question should resolve an ambiguity that affects coupling decisions. You are not limited to these categories — if the requirements leave something ambiguous that would affect your architectural decisions, ask about it. Ground questions in specific requirements you read.

- Business areas where core vs supporting vs generic classification is ambiguous — propose your interpretation and ask the user to confirm or correct - Organizational constraints that affect module boundaries (team ownership, deployment units, shared infrastructure) - Strategic direction that affects where volatility will be highest and where to invest design effort - Integration requirements where the appropriate coupling strength is unclear

  1. Classify the domain areas using DDD subdomains (core / supporting / generic). This determines volatility and where to invest design effort. Analyze the requirements and propose classifications yourself. Present them as a table:
SubdomainClassificationRationale
{area 1}Core{why}
{area 2}Supporting{why}
{area 3}Generic{why}

Then ask the user to validate using AskUserQuestion:

HeaderQuestionOptions
SubdomainsDo these subdomain classifications look right?1. Approved - All correct 2. Some are wrong - I'll tell you which to change 3. Missing subdomains - There are areas not listed

If the user says some are wrong, ask which ones and what the correct classification should be.

Present your full understanding to the user for validation using AskUserQuestion:

HeaderQuestionOptions
ApprovalDoes this understanding of the requirements look correct?1. Approved - Proceed to architecture design 2. Needs changes - I'll explain what's wrong 3. Missing context - There's more I should tell you

Do not proceed until approved.

Step 2: Design the Modular Architecture

Using the Balanced Coupling model:

  1. Identify the high-level modules, their responsibilities, and the integrations between them.
  2. Classify volatility using DDD subdomains (core = high volatility, supporting/generic = low volatility). This determines where to invest design effort.
  3. Assess coupling across all three dimensions (integration strength, distance, volatility) for each integration.
  4. Apply the balance rule: BALANCE = (STRENGTH XOR DISTANCE) OR NOT VOLATILITY

- For high-volatility components, design integrations so that strength and distance are balanced: - Components that must share high knowledge (functional/model coupling) should be co-located (low distance) -> high cohesion. - Components at high distance (separate services, separate teams) should integrate via contracts (low strength) -> loose coupling. - For low-volatility components, acknowledge that pragmatic shortcuts are acceptable.

  1. Flag unbalanced coupling: high strength + high distance (distributed monolith risk) or low strength + low distance (low cohesion / big ball of mud risk).
  2. Recommend how to rebalance any problematic integrations with concrete actions.

Present the coupling assessment table to the user:

IntegrationStrengthDistanceVolatilityBalanced?Action
A -> BModelHigh (separate services)High (core)No — tight couplingReduce strength: introduce contract via API

Work through each step with the user using AskUserQuestion. Each step requires user approval. Do not proceed to writing design documents until the modular architecture is fully validated by the user.

HeaderQuestionOptions
ApprovalDoes this modular architecture look correct?1. Approved - Proceed to design documents 2. Needs changes - I'll explain what to adjust 3. Rethink - Let's reconsider the module boundaries

Step 3: Write Module Design Documents

Using the validated architecture from Step 2, for each module create docs/design/!date +%Y-%m-%d/{module-name}/design.md containing:

# {Module Name}

## Functional Responsibilities

What this module does — the functionality it implements and the business capabilities it provides.

## Encapsulated Knowledge

What this module knows that no other module should — the domain concepts, business rules, and implementation details it owns.

## Subdomain Classification

Core / Supporting / Generic — and the rationale for the classification.

## Integration Contracts

For each module this one integrates with:

- **Direction**: Which module depends on which
- **Contract type**: The integration strength level (contract / model / functional)
- **What is shared**: The specific knowledge exchanged
- **Contract definition**: The interface, API, events, or data structures that define the boundary

## Change Vectors

Reasonable future changes that would require ONLY this module to change — the axes of evolution this module's boundary is designed to support.

Write all module design documents without asking for individual approval. The modular architecture was already approved in Step 2 — the documents are a direct translation of that approved design.

After writing all module documents, present the complete set to the user for review using AskUserQuestion:

HeaderQuestionOptions
ModulesAll module design documents have been written. How do they look?1. Approved - Proceed to test specifications 2. Needs changes - I'll explain which modules need work 3. Revisit architecture - The documents reveal a boundary problem

Iterate until approved.

Step 4: Write Module Test Specifications

For each module, create docs/design/!date +%Y-%m-%d/{module-name}/tests.md containing:

# {Module Name} — Test Specification

## Unit Tests

Tests for the module's internal logic in isolation. Covers business rules, calculations, state transitions, and edge cases.

## Integration Contract Tests

Tests that verify the module honors its integration contracts — that it produces the correct outputs given valid inputs according to its contract definitions.

## Boundary Tests

Tests that verify the module correctly rejects invalid inputs, handles edge cases at its boundaries, and maintains its encapsulation (nothing leaks).

## Behavior Tests

Tests that verify the module's functional responsibilities from an outside-in perspective — given a business scenario, the module behaves as expected.

Each test section should contain specific, named test cases with:

  • Test name: A descriptive name
  • Scenario: What is being tested
  • Expected behavior: What the correct outcome is

Write all test specifications without asking for approval. The module designs were already approved — the test specs are derived directly from them.

Step 5: Write the Architecture Document

Create docs/design/!date +%Y-%m-%d/architecture.md containing:

# Architecture Overview

## Functional Requirements Summary

Brief summary of the requirements this architecture addresses.

## Module Map

List of all modules with one-line descriptions.

## How the Modules Work Together

For each key functional flow / use case:

- Which modules participate
- How data/control flows between them
- What contracts govern the interactions

## Coupling Assessment

The coupling assessment table from the modular architecture analysis, with commentary on the key design decisions and their rationale grounded in the Balanced Coupling model.

## Design Decisions and Trade-offs

Key architectural decisions, what was considered, what was chosen, and why — grounded in the coupling dimensions and balance rule.

## Unresolved Risks

Anything the design intentionally leaves open, along with the conditions under which it should be revisited.

Write the architecture document without asking for approval. It synthesizes the already-approved module designs.

Step 6: Modularity Review

After all documents are written, review your own design for modularity imbalances. For each integration between modules:

  1. Map the integration: What knowledge is shared? Is it implicit or explicit? What is the integration strength level?
  2. Assess all three dimensions: strength, distance, and volatility.
  3. Apply the balance rule: BALANCE = (STRENGTH XOR DISTANCE) OR NOT VOLATILITY
  4. Flag imbalances: Focus on integrations that are both unbalanced and volatile.

For each issue found, classify its severity:

  • Critical: High strength + high distance + high volatility.
  • Significant: Unbalanced coupling in a moderately volatile area, or implicit coupling that hides integration points.
  • Minor: Unbalanced coupling in a low-volatility area, or low cohesion that increases cognitive load but doesn't cause cascading changes.

If there are any Critical or Significant issues:

  1. Present the issues to the user with the same structure used in step 2's coupling assessment table, plus a description of the knowledge leakage and recommended improvement for each.
  2. Propose concrete changes to the design to rebalance the coupling.
  3. Once the user approves the changes, update the affected design documents (module designs, test specs, and architecture document).
  4. Repeat this step — review the updated design again. Continue until no Critical or Significant issues remain.

Minor issues should be noted in the architecture document's "Unresolved Risks" section but do not block completion.

Important Constraints

  • Never assume. If anything is unclear, ask the user before proceeding.
  • Write actionable documents. Every design doc should give a developer enough context to start implementing. Every test spec should give a developer enough context to start writing tests.
  • Never recommend "just decouple everything." That ignores the balance between strength and distance.
  • Never evaluate coupling using only one dimension. Always consider all three: strength, distance, and volatility.
  • Distinguish essential from accidental volatility. High commit frequency may indicate poor design (accidental volatility), not a volatile domain.
  • Consider the organizational dimension of distance. Same code structure + different teams = higher effective distance.
  • Ground every recommendation in the model. Reference the specific dimension and principle that justifies each design decision.

平台分布

Codex

33.98%
按下载量换算39

Claude

31.66%
按下载量换算36

Cursor

21.66%
按下载量换算25

Gemini CLI

9.1%
按下载量换算10

安全审计

Gen Agent Trust Hub

通过

Socket

通过

Snyk

通过

权限和风险

执行命令

安装流程涉及命令执行,可能通过 npx skills add https://github.com/vladikk/modularity --skill design 联网下载 Skill 或依赖。用户安装前应确认命令来源、仓库内容和执行环境。

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。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