Token导航 LogoToken导航

Agent Skill · 研究检索

wtf.design-feature(wtf 设计特点)

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。

wtf.design-feature 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,权限为调用外部服务;许可证信息还不明确,暂无第三方扫描结果,按安装量在研究检索类的 73,675 个 Skill 里排第 56,238。

总安装 461周安装 19GitHub Stars 3下载 150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:wtf.design-feature(wtf 设计特点)
来源仓库:https://github.com/xiduzo/wtf
仓库路径:skills/wtf.design-feature
安装命令:
npx skills add https://github.com/xiduzo/wtf --skill wtf.design-feature
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/xiduzo/wtf --skill wtf.design-feature

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。

wtf.design-feature 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 方案或改进组件层级。使用时需结合现有品牌、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表现。

通过 npx skills add 命令从 GitHub 仓库安装,建议确认权限及是否会触发文件读写或构建流程。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。

来源仓库内路径:skills/wtf.design-feature

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/xiduzo/wtf --skill wtf.design-feature;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

    它需要调用外部服务,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

wtf.design-feature 怎么安装?

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

wtf.design-feature 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。

wtf.design-feature 需要什么权限?

权限级别是调用外部服务。该 Skill 可能调用第三方服务、云服务或外部模型 API,使用前需要确认账号、额度、数据发送范围和服务条款。

wtf.design-feature 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。

wtf.design-feature 用的人多吗?

公开数据是累计安装 461 次,近一周安装 19 次,下载 150 次,GitHub 3 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。

wtf.design-feature 安全吗?

没有查到第三方扫描结果,安全性无法从公开资料判断,安装前请自行审查源码、依赖和命令行为。

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。
  • 没有第三方扫描结果,安全性只能靠自己读源码判断。
  • 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

  • 需要明确授权才能用到公司项目里的场景。
  • 要求有安全扫描结论才肯安装的团队。
  • 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
wtf.design-feature

wtf 设计特点 · 本页

4613调用外部服务未明确未收录

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 Feature

Pick up a Feature as a designer and produce a holistic design covering the full user journey before tasks are cut. Core value: maps every user story to screens and states, identifies shared components across the feature, and writes a complete Design Handoff back into the Feature issue — so wtf.feature-to-tasks can derive better tasks and wtf.design-task inherits feature-level decisions rather than reinventing them.

Process

0. GitHub CLI setup

Run steps 1–2 of ../references/gh-setup.md (install check and auth check). Stop if gh is not installed or not authenticated. Extensions are not required for this skill.

Skip this step if invoked from another skill that already ran gh-setup this session.

1. Identify the Feature

If the user provided an issue number in their request, use it directly. Otherwise search for recent open issues with label feature to populate options and call AskUserQuestion with question: "Which Feature are you designing?", header: "Feature", and options pre-filled with 1–2 likely open Feature issue references.

Fetch the Feature and its parent Epic:

gh issue view <feature_number>    # User stories, ACs, Edge Cases, Domain Events
# Extract epic number from Context section, then:
gh issue view <epic_number>       # Goal, Context, Design Artifacts (strategic input)

Extract and hold in context:

  • Feature capability name (Actor + verb + object)
  • All user stories
  • All Acceptance Criteria
  • Edge Cases
  • Domain Events (emitted/consumed)
  • Epic's Design Artifacts (any Figma links or research docs — these are upstream constraints)

2. Lifecycle check

Check whether the feature already has a designed label:

gh issue view <feature_number> --json labels --jq '.labels[].name'

If designed is present, call AskUserQuestion with:

  • question: "This Feature already has a designed label. Continuing will overwrite the existing Design Handoff. How would you like to proceed?"
  • header: "Already designed"
  • options: [{label: "Redesign it", description: "Overwrite the existing Design Handoff with a new one"}, {label: "Exit", description: "Leave the existing design as-is"}]
  • Redesign it → continue.
  • Exit → exit immediately.

If absent, continue silently.

3. Load the design steering document

Use the Read tool to attempt reading docs/steering/DESIGN.md.

If it exists: keep its content in context. Apply its design principles, tokens, component patterns, and accessibility standards silently throughout this session.

If it does not exist, call AskUserQuestion with:

  • question: "docs/steering/DESIGN.md doesn't exist yet. This document captures your design principles, tokens, and component patterns. Would you like to create it now?"
  • header: "Design steering doc missing"
  • options: [{label: "Create it now", description: "Run steer-design before continuing (recommended)"}, {label: "Skip for this session", description: "Continue without it — design decisions won't reference project standards"}]
  • Create it now → follow the wtf.steer-design process, then return here and continue from step 4.
  • Skip for this session → continue without it.

4. Explore the design system and codebase

Use the Agent tool with these searches (run in parallel):

  • Glob('src/components/**/*', 'src/**/components/**/*', 'components/**/*') — existing UI components; flag any that map to domain objects in this Feature's user stories
  • Glob('**/{tokens,theme,variables,design-tokens}.{css,scss,ts,js,json}') — design tokens
  • Glob('src/**/*.{stories,story}.{ts,tsx,js,jsx,mdx}') — Storybook stories as pattern references for similar flows
  • Grep for figma.com URLs across .md, .mdx files — existing Figma references in related issues or docs

Note which existing components can be reused vs which are new. This feeds step 7.

5. Map the full user journey

For each user story ("As a [Actor], I want [action] so that [outcome]"), derive:

  1. Entry point — what triggers this story? (screen, action, event)
  2. Happy path screens — ordered list of screens/states the actor moves through
  3. Branch states — loading, error, empty, partial data, permission denied
  4. Exit point — what confirms the story is complete for the actor?

Also map:

  • Cross-story transitions — screens shared between multiple user stories
  • Edge case screens — one screen per Edge Case from the Feature issue
  • Domain Event surfaces — where in the UI does each emitted Domain Event become visible to the actor?

Produce a journey map as a structured list — do not ask the user, derive from user stories and ACs.

6. Ask about design assets

Call AskUserQuestion with question: "How would you like to handle designs for this feature?", header: "Design assets", and options:

  • {label: "I have Figma frames", description: "Provide frame URLs — I'll validate coverage against the full journey map"}
  • {label: "Generate designs for me", description: "I'll use Figma MCP to generate frames from the user stories and design system"}
  • {label: "Scaffold a brief only", description: "No Figma — produce a text screen inventory and component map"}
  • {label: "Partial — some screens designed", description: "Provide available frames; remaining screens go to generate or scaffold"}

Path A — Human provides frames: Collect the top-level Figma file URL plus individual frame URLs. For each screen in the journey map (step 5), check whether a frame covers it. Present a coverage matrix: screen → frame URL (or ⚠ gap). If gaps exist, call AskUserQuestion asking whether to generate missing frames (Path B) or leave them as pending.

Also validate provided frames against spec:

  • Every user story has at least one matching frame
  • Every edge case from the Feature issue has a matching error/boundary state frame
  • Every Domain Event surface identified in step 5 is represented

Flag any validation failures as gaps in the coverage matrix.

Path B — AI generates via Figma MCP: Check whether the Figma MCP tool generate_figma_design is available. If unavailable, warn the user and fall back to Path C (scaffold).

If available: for each screen in the journey map without a frame, call generate_figma_design with:

  • The screen's user story and entry/exit points as the design brief
  • Component patterns and tokens from docs/steering/DESIGN.md (loaded in step 3)
  • Shared components identified in step 7 as reuse constraints
  • Any Figma URLs from the Epic's Design Artifacts as style reference

Collect the generated frame URLs and treat them as Path A frames for the coverage matrix and Design Handoff.

Path C — Scaffold brief only: For each screen in the journey map, produce a text brief listing required UI elements, interactions, and relevant design tokens. This is a Figma-free design brief a designer or developer can execute against. Use references/component-spec-template.md as the structure if available.

Partial: Collect available frame URLs, run Path A validation on covered screens. For uncovered screens, call AskUserQuestion asking whether to generate (Path B) or scaffold (Path C) the remainder.

7. Identify shared components

Across all screens in the journey map, identify:

  • Reused existing components — already in the codebase (found in step 4); list component path + which screens use it
  • New shared components — appear on 2+ screens but do not exist yet; name them using domain language
  • Screen-specific components — appear on only one screen; note them but do not detail here (that is wtf.design-task's job)

This component map reduces duplication when wtf.design-task runs per-task.

8. Draft the Design Handoff

Produce content for the Design Handoff section of the Feature issue:

## Design Handoff

- Figma: <top-level Figma file URL, generated file URL, or "pending (scaffold only)">
- Flow: <link to prototype/flow if available>
- Design path: <Path A: human-provided | Path B: AI-generated | Path C: scaffold brief>

### Screen inventory

| Screen | Story | Figma frame | States covered | Source |
|--------|-------|-------------|----------------|--------|
| <screen name> | As a... | <url or pending> | default / loading / error / empty | provided / generated / scaffolded |

### Validation (Path A/B only)

- [ ] Every user story has ≥1 frame
- [ ] Every edge case has a boundary/error state frame
- [ ] Every Domain Event surface is represented
- [ ] All frames consistent with DESIGN.md tokens and patterns

### Shared components

| Component | Exists? | Path or new | Used on |
|-----------|---------|-------------|---------|
| <name> | yes/no | <path or "new"> | <screens> |

### Accessibility notes

<any feature-level a11y constraints from steering doc or Epic>

### Open gaps

<list any screens or states not yet designed — pending Figma frames>

9. Review with user

Show the draft. Then call AskUserQuestion with question: "Does this cover the full feature journey?", header: "Review", and options:

  • {label: "Looks complete — update the issue", description: "Proceed"}
  • {label: "Missing screens or states", description: "I want to add coverage"}
  • {label: "Other changes", description: "I want to adjust something"}

Apply edits, then proceed.

10. Update the Feature issue

Read the current issue body, replace only the Design Handoff section with the new content, preserve all other sections. Write to temp file and use --body-file:

gh issue edit <feature_number> --body-file /tmp/wtf.design-feature-<feature_number>-body.md

Add the designed label when either:

  • Path A/B: all screens in the journey map have Figma frames (no open gaps in coverage matrix), or
  • Path C: the full scaffold brief is complete (no Figma expected)
gh issue edit <feature_number> --add-label "designed"

If Path A/B has open gaps, do not add designed — note it will be added once gaps are closed.

This fulfills the Feature DoR gate: "Design handoff complete".

Print the updated Feature issue URL.

11. Offer to continue

Call AskUserQuestion with:

  • question: "What's next?"
  • header: "Next step"
  • options:

- {label: "Break into Tasks", description: "Run feature-to-tasks — design context will inform task breakdown (default)"} - {label: "Design another Feature", description: "Design another Feature for the same Epic"} - {label: "Stop here", description: "Exit — no further action"}

  • Break into Tasks → follow the wtf.feature-to-tasks skill, passing the Feature number in as context. Note to the user that wtf.design-task will inherit the shared component map from this Design Handoff.
  • Design another Feature → restart from step 1, reusing the same Epic context.
  • Stop here → exit.

平台分布

Codex

39.14%
按下载量换算59

Claude

27.54%
按下载量换算41

Cursor

17.61%
按下载量换算26

Gemini CLI

8.82%
按下载量换算13

安全审计

暂无安全审计结果可展示。

权限和风险

调用外部服务

该 Skill 可能调用第三方服务、云服务或外部模型 API,使用前需要确认账号、额度、数据发送范围和服务条款。

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