界面已本地化;技能介绍与说明文档按所选语言由 AI 自动翻译,代码、命令、路径和专业术语保留原文,以原作者文档为准。
探索技能
设计与创意 · SKILL.md

redesign

redesign

在不破坏功能的情况下将现有网站或应用升级为高端质量——审计当前设计,识别通用/AI 特征,然后以外科手术式方式应用品味和系统规则。当用户想改进、现代化、润色或“make better”现有 UI/代码库时使用。

AI 翻译

独立 GitHub 快照 · 自动分类 · 以原作者文档为准

技能说明

GitHub

name: redesign description: Upgrade an existing website or app to premium quality without breaking functionality — audit the current design, identify generic/AI tells, then apply taste and system rules surgically. Use when the user wants to improve, modernize, polish, or "make better" an existing UI/codebase.

invocation: was not a field Claude Code reads - unknown frontmatter keys are

ignored without an error, so these six were auto-invocable the whole time they

were marked user-only. This is the field that actually does it.

disable-model-invocation: true

Skill: Redesign & Audit

Step 0 — is the kit here? This skill reads files from the kit. Check once: ls ${CLAUDE_SKILL_DIR}/../../../tokens >/dev/null 2>&1 && echo KIT_OK || echo KIT_MISSING On KIT_MISSING only the skill folders were installed, which is what npx skills add does. Say so plainly, point the user at npx ux-ui-agent-skills init or the plugin install, and stop. Do not guess the contents of a file you could not open.

Audit-first redesign that preserves behavior.

Steps

  1. Read ${CLAUDE_SKILL_DIR}/../../../workflows/redesign-audit.md (the full Scan → Diagnose → Direct → Apply → Verify sequence + Output Completeness).
  2. Scan the codebase: framework, styling method (→ route to the matching frameworks/ adapter), token reality, components, what must be preserved.
  3. Diagnose with the Banned Defaults checklist (${CLAUDE_SKILL_DIR}/../../../taste/design-taste.md) + the review rubric (design-review skill). Produce a prioritized findings table.
  4. Direct: choose an archetype/system (apply-aesthetic skill) that fits the brand.
  5. Apply in order — tokens first, then typography/spacing, then component states, then motion — without changing routes/data/markup semantics (except a11y fixes).
  6. Verify: re-run design-review + a11y-audit; smoke-test every previously working flow; dark mode + responsive spot-check. Run ${CLAUDE_SKILL_DIR}/../../../scripts/validate_contrast.py on the token source and ${CLAUDE_SKILL_DIR}/../../../scripts/lint_hardcodes.py over the changed code.

Guardrails

Never sacrifice a working feature for aesthetics. Never ship a brand color that fails contrast. Never remove existing accessibility affordances. Deliver complete files — no placeholders.

Single-theme consistency (critical for multi-page apps)

Consolidate to ONE shared token theme and make every page consume it — a redesign that leaves different pages on different palettes has failed. Replace per-page/ad-hoc colors with semantic tokens; verify with ${CLAUDE_SKILL_DIR}/../../../scripts/lint_hardcodes.py that no page reintroduces off-theme values. Theme switches must come from the single token source, not page edits (.claude/rules/tokens-and-color.md → Single-Theme Consistency).

我的收藏