将作品集网站路由与桌面和移动尺寸下的批准的视觉参考进行比较，然后报告有证据支持的视觉缺陷和修复方案。
该技能用于对个人作品集网站进行视觉验证，而非仅进行代码审查。其核心工作流程包括：搭建环境——确认参考、路由、UI状态、视口尺寸及任务是否允许实施；比较——从层级、排版、间距、色彩、资源保真度、响应式裁剪/溢出及交互状态等方面进行评估，并检查键盘可达性和减少动效行为；报告——首先说明证据和限制，然后按优先级（P1:阻塞核心任务、P2:明显视觉或响应式不匹配、P3:可延后的修饰）对发现进行分组，每个发现需指明受影响的路由/状态、观察到的行为、可能根源以及最小的可验证修复方案。该技能特别强调在浏览器实际渲染输出与参考图之间进行视觉比对，而非静态代码检查。
npx skills add https://github.com/husheng-web/personal-website-skills/tree/HEAD/skills/portfolio-visual-qa执行视觉验证，而非仅进行代码审查。
确认参考、路由、UI 状态、视口尺寸以及任务是否允许实施。通过其既定命令启动网站，并在请求的尺寸下检查渲染输出；如果请求未指定，应包括一个移动尺寸。
评估层级、排版和换行、间距和几何、色彩和对比度、资源保真度、响应式裁剪/溢出以及交互状态。当涉及动效或导航时，验证键盘可达性和减少动效行为。
首先说明证据和限制。按优先级对发现进行分组：
P1：阻塞预期的层级、访问或核心任务。P2：明显的视觉或响应式不匹配。P3：可延后的修饰。对于每个发现，指明受影响的路由/状态、观察到的行为、可能根源以及最小的可信修复方案。在受影响的视口中重新测试修复。当浏览器截图或目标参考不可用时，不得声称视觉检查通过。
This repo contains 7 individual skills — each has its own dedicated page.
Validate Chinese and English portfolio routes, content states, metadata, and fallbacks. Use when adding or reviewing localized portfolio content.
Build or update an evidence-backed portfolio case study in an existing website. Use after narrative facts and media have been approved.
Audit raw portfolio-project materials into an evidence-backed publishing brief. Use before writing a case study, project card, or factual portfolio copy.
Prepare, optimize, and register portfolio images and documents while preserving provenance, accessibility, and responsive rendering contracts.
Review or implement portfolio UI motion against an existing motion system, including reduced-motion, keyboard, performance, and responsive constraints.
Validate a statically exported portfolio website for GitHub Pages release readiness. Use before publishing or diagnosing deployment regressions.
Add or review an accessible portfolio project-preview reader for ordered long images, document pages, or staged visual modules.
ElevenLabs text-to-speech with mac-style say UX.
Oracle CLI second-model review/debug/refactor/design with selected files, dry-run token checks, API or browser engine.
Capture and automate macOS UI with the Peekaboo CLI.
Generate multiple radically different interface designs for a module using parallel sub-agents. Use when user wants to design an API, explore interface options, compare module shapes, or mentions "design it twice".
Build a throwaway prototype to flesh out a design — a runnable terminal app for state/business-logic questions, or several radically different UI variations toggleable from one route.
Interview the user relentlessly about a plan or design until reaching shared understanding, resolving each branch of the decision tree. Use when user wants to stress-test a plan, get grilled on their design, or mentions "grill me".