Sdlc Ui Visual

Community

TsCarpe

sdlc-ui-visual

Design-to-code visual restoration loop: diffs the running page against a Figma link or baseline image via real screenshots (DOM geometry + pixel diff/heatmap + zoomed inspection), drives tiered fix rounds and verifies functional regression. 设计稿到代码的视觉还原闭环：以 Figma 链接或基准图为唯一视觉基准，对运行中的前端页面做浏览器截图比对（DOM 几何 + 像素差异/热区 + 小元素放大），按 P0>P1>P2 分级驱动修复轮，功能回归后按六要素汇报。Use when 用户提供 Figma 设计稿或基准图要求视觉还原、UI 还原、设计稿还原、照着 Figma 改样式、像素级比对、视觉走查、视觉偏差修复，或说 ui check、ui-align。输入是设计稿基准且允许改样式代码——纯功能验证走 sdlc-test。用法：/sdlc-ui-visual Figma链接或基准图路径 [目标页面或组件]

Overview

Publisher	TsCarpe
Repository	`claude-sdlc-skills`
Skill name	`sdlc-ui-visual`
Stars	65
Forks	0
Bundled files	13
License	MIT
Links	Skill source Repository

Markdown instructions
A SKILL.md file the model loads on demand, so it only costs tokens when a request actually matches.
Works with any LLM
AI skills are plain Markdown, not provider-specific code, so this works with GPT, Claude, Gemini, Grok, or a local model.
13 bundled files
Scripts, templates, and references the model can read while it works. Files are read-only and never executed.
Open source
Published by TsCarpe on GitHub. Read the source before you install it.

Installation

Install the Sdlc Ui Visual AI skill in TypingMind to use it with any LLM, or drop it into another agent that reads SKILL.md.

Install in TypingMind

TypingMind installs a skill straight from its GitHub folder — it reads SKILL.md, bundles the resource files, and stores the result locally.

Open the app and go to Plugins → Skills.
Choose "Install from GitHub".
Paste the skill folder URL below and confirm.
https://github.com/TsCarpe/claude-sdlc-skills/tree/main/skills/sdlc-ui-visual
Enable the skill in any chat where you want it available.

Plugins → Skills → Add skill → From GitHub URL, then paste the folder URL and press Continue.

Install in another agent

Any agent that reads the Agent Skills format can use this skill — copy the folder into that agent's skills directory.

Claude Code — .claude/skills

git clone --depth 1 https://github.com/TsCarpe/claude-sdlc-skills.git /tmp/claude-sdlc-skills
mkdir -p .claude/skills
cp -r /tmp/claude-sdlc-skills/skills/sdlc-ui-visual .claude/skills/sdlc-ui-visual

Restart Claude Code after copying so it picks up the new skill.

Use it in TypingMind

Enable Sdlc Ui Visual in any TypingMind chat and the model takes it from there. Its name and description sit in the system prompt, and the moment a request matches, the model loads the full instructions itself — you never invoke it by hand, and it costs no tokens until it is actually used.

The model loads Sdlc Ui Visual on its own as soon as a request matches it.

Works with any AI model

AI skills are plain Markdown instructions rather than provider-specific code, so Sdlc Ui Visual is not tied to the model it was written for. Install it once in TypingMind and use it with GPT-5, Claude, Gemini, Grok, DeepSeek, Mistral, Llama, or a local model you run yourself — all on your own API keys.

Loaded only when it is needed
The system prompt carries just the name and description. The instructions are fetched on the first matching request, so an idle skill costs nothing.
Switch models mid-chat
Because the skill is instructions rather than code, changing model does not break it — the next model reads the same SKILL.md.

Skill instructions

This is the SKILL.md content the model loads. Read it before installing — a skill is instructions your model will follow.

UI 视觉还原

以设计稿为基准把运行中页面还原到位的闭环：基准 → 代码对照 → 稳定截图 → 多道比对 → 分级修复轮 → 功能回归 → 汇报。与 sdlc-test 分界：那边验「行为对不对」（源码只读），这边验「长得像不像设计稿」（以设计稿为基准、允许改样式代码）。

栈适配：方法论栈无关；Vue2 + element-ui 栈姿势见 references/stack-vue2-element.md（换栈先读，日常执行不加载）。

前置依赖：python3 + Pillow（compare.py；缺依赖时脚本会提示 pip3 install Pillow）；figma-context-mcp（Figma 输入，server 别名以实际安装为准）；浏览器自动化 MCP（chrome-devtools，不可用时按 pipeline 降级链切 Playwright）。

产物目录（相对目标项目根）：.ui-check/<目标名>/——独立于 sdlc/ 工作区（无 sdlc 工作区也成立，有则互不依赖）。目录内二进制产物多，首次创建时提示用户把 .ui-check/ 加入 .gitignore。

状态落盘（run.md）：.ui-check/<目标名>/run.md 是唯一状态源——七步进度、差异分类台账、待用户问题、已拍板记录全在其中，每步结束即更新；中断后重开会话 = 读 run.md 头部进度续跑，禁止凭对话记忆恢复、禁止推翻台账中已拍板分类。模板见 references/templates.md。

步骤 0：环境与可达性

浏览器会话可用、目标路由可达、登录态有效。遇登录墙（SSO 等）不自行绕：请用户人工完成登录与页面导航，之后截图/断言/比对全自动化（人工分工要记入 run.md）。dev server 先探测再决定是否启动；浏览器工具冲突时按 references/pipeline.md 降级链切换。

步骤 1：基准获取

Figma 链接：/design/ 后第一段为 fileKey；URL 里 node-id=12-3456 的 - 转 : 得 12:3456 作 nodeId。figma-context-mcp:get_figma_data 拿结构化数据（输出过大时持久化本地再解析），figma-context-mcp:download_figma_images 下载基准 PNG（pngScale=2，比对前等比缩到 1x，禁止拉伸；MCP server 别名以实际安装为准，如 figma-context-mcp）。
仅基准图：跳过结构化数据，尺寸/颜色靠图像分析与人工读数。
frame 选择策略（页面级）：node-id 即整页 frame（只用于上下文理解——应用壳与实际项目不同，整页比对无意义）；比对基准 = 主内容区 frame：从子 frame 按尺寸启发式筛（宽度最接近页面主体宽度者）+ 唯一性校验，多候选/零候选暂停问用户。组件级任务（弹窗等）node-id 本身即基准。选择结果记入 run.md。
期望值清单：从基准数据建 .ui-check/<目标名>/expected.md——每元素位置/尺寸/字号/字重/颜色/圆角/间距，<20px 的小元素逐项打标 小元素⚠（步骤 4 强制元素级验证）。细则见 references/pipeline.md。

步骤 2：代码对照（修复动作前必做）

两个确定产物，不扩张成全量读码：

git diff 已改样式盘点：任务常发生在开发已写初版之上，先看未提交改动里已对齐了什么
全局样式对抗清单：grep 全局覆盖文件（主题/变量/组件库全局 less 等），列出会压制本目标样式的 !important 与全局规则——不盘点就写样式 = 写无效样式

步骤 3：浏览器稳定化 + 截图

视口 = 设计稿画布尺寸，保持比例一致。每轮截图前稳定化三件（缺一截图无效，细则见 references/pipeline.md）：

轮询等主内容挂载（页面根 class 出现）
关闭干扰弹窗——弹窗不关 = 截图是旧帧且与 DOM 不一致
等数据加载（树节点/列表行数 > 0）再截

锚定对齐：getBoundingClientRect() 实测参照元素位置作为锚点，设计稿内容区 (0,0) 对齐到该点，不做全页对齐。截图存轮次目录。

步骤 4：多道比对

bash
python3 <skill 目录>/scripts/compare.py --actual <截图> --design <基准图> \
  [--ox <锚点x> --oy <锚点y>] [--box x y w h] --outdir .ui-check/<目标名>/r<N>/

产出：50% 叠加图 / 差异图 / 热区网格（附格中心坐标）/ mean diff / >30 差异像素占比 / 暗像素计数，stats 落盘轮次目录。

第 1 道 DOM 几何：期望值清单逐项对照 getBoundingClientRect + getComputedStyle（精确到 0.1px）
第 2 道像素比对：叠加 + 差异图 + 热区，抓几何对照漏掉的渲染问题
小元素强制验证（独立于上面两道的结果）：凡期望值清单打标 小元素⚠ 的元素，一律 --box 裁剪放大 ≥3 倍 + 视觉判读「形状/大小/颜色」+ 暗像素计数——两道全过也不豁免（区域平均会稀释小元素差异，数值等价 ≠ 视觉等价）

比对产物的读法与「不可判定 → 升级验证手段」路径见 references/triage.md。

步骤 5：差异分类 + 修复轮

先按 references/triage.md 决策表分类（结构性/组件级 = 修；内容性/需求性/应用壳层 = 保留并说明；功能性新元素 = 问用户；共用组件 = 权衡调用方影响），分类结果登记 run.md 台账，再修：

优先级 P0 整体布局/尺寸/溢出 > P1 模块间距/字体/对齐 > P2 颜色/边框/圆角/图标
先修父容器与整体布局，再抠组件细节；禁止叠加 margin / 硬编码坐标 / 绝对定位掩盖根因
≤5 轮；「无改善」= mean diff 与 >30 差异像素占比两个指标连续 2 轮均不降 → 停止修改，重新分析根因（重读期望值清单 + DOM 复测 + 必要时问用户），不盲改

步骤 6：功能回归

视觉让位功能，修复不得破坏业务：

交互断言（按钮/表单/弹窗/菜单，内联断言一次跑完，模板见 references/pipeline.md）
同页面其他 Tab/状态回归 + 共用组件的其他调用方页面回归（改共用组件必查）；回归截图与结论统一落 本目标 .ui-check/<目标名>/regression/（调用方页面不单开目录，保证台账可核查）
lint + console 增量逐条归因（与本次改动无关才放行）+ 编译产物检查

步骤 7：汇报

按 references/templates.md 六要素模板：改动文件与原因 / 比对轮次与数据（引用 stats.txt 原值与产物路径）/ 量化改善 / 产物位置 / 回归与 lint 结果（含未验证项明列，无则写「无」）/ 保留的偏差清单（偏差/原因/是否需人工确认）。

红线

不得假称完成比对：环境不支持截图/工具全不可用时，明确声明限制并停在能落证据的步骤，禁止凭代码推测编造比对结论
不为视觉匹配破坏业务：不隐藏元素、不删内容、不改交互语义；用 display:none 消灭差异 = 作弊
设计稿是唯一视觉基准，不擅自修改替换基准图；超范围项（应用壳/功能性新元素）问用户后跳过并留档
用户模糊反馈（「太靠左」）先像素级分析其截图，再用选择题确认期望方向，避免改反
栈特有坑（组件库对抗/指令陷阱/图标替换）动手前查 references/stack-vue2-element.md；栈无关坑查 references/pitfalls.md

反合理化

"代码看起来对了" → 无截图不结论，必须真实截图验证
"区域平均 diff 已到噪声水平，小图标不用抠了" → 小元素⚠ 打标项强制元素级验证，两道全过也不豁免
"这轮没改善，再改一轮试试" → 双指标连续 2 轮不降即停，重新分析根因
"用户说按钮太小，但 API 数据是 10×10 一致的" → 数值等价 ≠ 视觉等价，做暗像素计数与放大判读再下结论
"弹窗挡着，先截了再说" → 不关弹窗截图是旧帧，稳定化三件缺一无效

Bundled files

The model reads these on demand while the skill is loaded. They are exposed as readable files and are never executed.

Frequently asked questions

What does the Sdlc Ui Visual AI skill do?

Why use Sdlc Ui Visual on TypingMind?

Because you install it once and use it with any model. Sdlc Ui Visual is plain Markdown rather than provider-specific code, so the same skill runs on GPT-5, Claude, Gemini, Grok, or a local model — and you can switch model mid-chat without it breaking. TypingMind runs on your own API keys, so you pay providers directly instead of a per-seat subscription, and your skills and chats stay in your own storage.

How do I install Sdlc Ui Visual in TypingMind?

Open Plugins → Skills → Install from GitHub in TypingMind and paste https://github.com/TsCarpe/claude-sdlc-skills/tree/main/skills/sdlc-ui-visual. TypingMind reads its SKILL.md and bundles its files and installs it as a skill you can enable per chat.

Which AI models can use Sdlc Ui Visual?

Any model you connect in TypingMind. AI skills are plain Markdown instructions rather than provider-specific code, so GPT, Claude, Gemini, Grok, and local models can all load this skill when a request matches it.

How many AI models can I use with Sdlc Ui Visual?

As many as you like. As long as a model supports skills, you can use Sdlc Ui Visual with it — GPT, Claude, Gemini, Grok, DeepSeek, Mistral, Llama and more — all on TypingMind with your own API keys.

Is the Sdlc Ui Visual AI skill free?

Yes. It is published on GitHub by TsCarpe under the MIT license. You only pay your own AI provider for the tokens you use.

What are AI skills?

An AI skill is a reusable instruction bundle that teaches an AI model how to do one specific task. It follows the open Agent Skills format: a SKILL.md file with a name and description, plus any scripts, templates or reference files the model may need. The model reads the instructions only when your request matches the skill, so an installed skill costs nothing until it is used.

How are AI skills different from plugins or MCP servers?

A plugin or MCP server gives a model new tools to call — code that runs somewhere and returns a result. An AI skill gives the model knowledge and process instead: how to approach a task, which steps to follow, what good output looks like. Skills are plain Markdown, so they need no server, no API key and no runtime, and they work with any model.

View all

Release Openclaw Maintainer

openclaw

OrganizationPopular

Prepare, publish, recover, or verify OpenClaw beta, stable, and extended-stable releases, including approved backports.

Release Openclaw Plugin Testing

openclaw

OrganizationPopular

Plan and run pre-release OpenClaw plugin validation across bundled plugins, package artifacts, lifecycle commands, doctor/fix, config round-trip, gateway startup, SDK compatibility, Docker E2E,…

Solid

Write, port, review, or test Solid 2 UI in OpenClaw (Control UI, plugin views, Canvas hosts), including state ownership, Lit interop during the migration, native overlays, testing, performance, and…

Auto Qa

Run OpenClaw-wide autonomous QA and live/stress campaigns across independent subsystem lanes, with verified fixes and a resumable evidence report.

Overview

Installation

Install in TypingMind

Install in another agent

Use it in TypingMind

Works with any AI model

Skill instructions

UI 视觉还原

步骤 0：环境与可达性

步骤 1：基准获取

步骤 2：代码对照（修复动作前必做）

步骤 3：浏览器稳定化 + 截图

步骤 4：多道比对

步骤 5：差异分类 + 修复轮

步骤 6：功能回归

步骤 7：汇报

红线

反合理化

Bundled files

Frequently asked questions

Related AI Skills

Set up your own AI workspace now