Codex compatibility note:
- Invoke repository skills with
$skill-name in Codex; this mirrored copy rewrites legacy Claude /skill-name references.
- Task tracker mandate: BEFORE executing any workflow or skill step, create/update task tracking for all steps and keep it synchronized as progress changes.
- User-question prompts mean to ask the user directly in Codex.
- Ignore Claude-specific mode-switch instructions when they appear.
- Strict execution contract: when a user explicitly invokes a skill, execute that skill protocol as written.
- Subagent authorization: when a skill is user-invoked or AI-detected and its protocol requires subagents, that skill activation authorizes use of the required
spawn_agent subagent(s) for that task.
- Do not skip, reorder, or merge protocol steps unless the user explicitly approves the deviation first.
- For workflow skills, execute each listed child-skill step explicitly and report step-by-step evidence.
- If a required step/tool cannot run in this environment, stop and ask the user before adapting.
Codex Project-Reference Loading (No Hooks)
Codex uses static project-reference loading instead of runtime-injected project docs.
When coding, planning, debugging, testing, or reviewing, open project docs explicitly using this routing.
Always read:
docs/project-config.json (project-specific paths, commands, modules, and workflow/test settings)
docs/project-reference/docs-index-reference.md (routes to the full docs/project-reference/* catalog)
docs/project-reference/lessons.md (always-on guardrails and anti-patterns)
Missing/stale context route: If docs/project-config.json, the docs index, lessons.md, CLAUDE.md, AGENTS.md, or any task-required reference doc is missing or stale, auto-run $project-init or the narrow setup route ($project-config, $docs-init, $scan-all, $scan --target=<key>, $claude-md-init) before ordinary project-specific work. If Codex mirrors or AGENTS.md are missing/stale, ask the user to run $sync-codex; do not auto-run it.
Situation-based docs:
- Project structure/architecture/tech-stack/deployment/setup (any layer — backend, frontend, or infra):
project-structure-reference.md
- Backend/CQRS/API/domain/entity changes:
backend-patterns-reference.md, domain-entities-reference.md
- Frontend/UI/styling/design-system:
frontend-patterns-reference.md, scss-styling-guide.md, design-system/README.md
- Spec authoring,
docs/specs/ pathing, or TC format: feature-spec-reference.md, spec-system-reference.md, spec-principles.md
- Behavior/public-contract changes or spec-test-code sync:
workflow-spec-test-code-cycle-reference.md plus the spec docs above
- Derived spec indexes/ERDs/reimplementation guides:
spec-system-reference.md and source Feature Specs under docs/specs/
- Integration test implementation/review:
integration-test-reference.md
- E2E test implementation/review:
e2e-test-reference.md
- Code review/audit work:
code-review-rules.md plus domain docs above based on changed files
Do not read all docs blindly. Start from docs-index-reference.md, then open only relevant files for the task.
Quick Summary
Goal: Provide searchable UI/UX design intelligence -- 50 styles, 21 palettes, 50 font pairings, 20 charts, 8 stacks.
Workflow:
- Analyze — Extract product type, style keywords, industry, and stack from user request
- Search — Query
search.py across domains: product, style, typography, color, landing, chart, UX
- Stack Guidelines — Get stack-specific best practices (default: html-tailwind)
- Implement — Synthesize search results into cohesive design and code
Key Rules:
- No emojis as UI icons (use SVG: Heroicons, Lucide, Simple Icons)
- All clickable elements need
cursor-pointer and hover feedback
- Light mode text must have 4.5:1 minimum contrast ratio
- Every async surface MUST handle all its states — show a loading indicator while in-flight, a user-visible error (with retry) on failure, and a meaningful empty state when there's no data. Never a frozen blank, a silent failure, or a blank list.
- Responsive by default — layouts MUST stay usable on small devices. Preferred: reflow (rows
flex-wrap or row → column, grids collapse to one column). Acceptable fallback when a layout genuinely can't reflow (tables, canvases, wide grids): fixed min-width/min-height + overflow: auto scroll — scrolling is OK. Hard minimum: nothing broken — no clipped, cut-off, or unreachable content. If small-screen support needs a refactor too big for the task, confirm scope with the user first.
- Test both light/dark modes, all UI states (loading/error/empty), and responsive breakpoints (320/768/1024/1440) before delivery
Pre-read (project design system): When the project has a design system, load designSystem.canonicalDoc + tokenFiles from docs/project-config.json and prefer project tokens/components over the generic style intelligence database.
Be skeptical. Apply critical thinking, sequential thinking. Every claim needs traced proof, confidence percentages (Idea should be more than 80%).
UI/UX Pro Max - Design Intelligence
Searchable database of UI styles, color palettes, font pairings, chart types, product recommendations, UX guidelines, and stack-specific best practices.
Prerequisites
Check if Python is installed:
python3 --version || python --version
If Python is not installed, install it based on user's OS:
macOS:
brew install python3
Ubuntu/Debian:
sudo apt update && sudo apt install python3
Windows:
winget install Python.Python.3.12
How to Use This Skill
When user requests UI/UX work (design, build, create, implement, review, fix, improve), follow this workflow:
Step 1: Analyze User Requirements
Extract key information from user request:
- Product type: SaaS, e-commerce, portfolio, dashboard, landing page, etc.
- Style keywords: minimal, playful, professional, elegant, dark mode, etc.
- Industry: healthcare, fintech, gaming, education, etc.
- Stack: React, Vue, Next.js, or default to
html-tailwind
Step 2: Search Relevant Domains
Use search.py multiple times to gather comprehensive information. Search until you have enough context.
python .claude/skills/ui-ux-pro-max/scripts/search.py "<keyword>" --domain <domain> [-n <max_results>]
Recommended search order:
- Product - Get style recommendations for product type
- Style - Get detailed style guide (colors, effects, frameworks)
- Typography - Get font pairings with Google Fonts imports
- Color - Get color palette (Primary, Secondary, CTA, Background, Text, Border)
- Landing - Get page structure (if landing page)
- Chart - Get chart recommendations (if dashboard/analytics)
- UX - Get best practices and anti-patterns
- Stack - Get stack-specific guidelines (default: html-tailwind)
Step 3: Stack Guidelines (Default: html-tailwind)
If user doesn't specify a stack, default to html-tailwind.
python .claude/skills/ui-ux-pro-max/scripts/search.py "<keyword>" --stack html-tailwind
Available stacks: html-tailwind, react, nextjs, vue, svelte, swiftui, react-native, flutter
Search Reference
Available Domains
| Domain |
Use For |
Example Keywords |
product |
Product type recommendations |
SaaS, e-commerce, portfolio, healthcare, beauty, service |
style |
UI styles, colors, effects |
glassmorphism, minimalism, dark mode, brutalism |
typography |
Font pairings, Google Fonts |
elegant, playful, professional, modern |
color |
Color palettes by product type |
saas, ecommerce, healthcare, beauty, fintech, service |
landing |
Page structure, CTA strategies |
hero, hero-centric, testimonial, pricing, social-proof |
chart |
Chart types, library recommendations |
trend, comparison, timeline, funnel, pie |
ux |
Best practices, anti-patterns |
animation, accessibility, z-index, loading |
prompt |
AI prompts, CSS keywords |
(style name) |
Available Stacks
| Stack |
Focus |
html-tailwind |
Tailwind utilities, responsive, a11y (DEFAULT) |
react |
State, hooks, performance, patterns |
nextjs |
SSR, routing, images, API routes |
vue |
Composition API, Pinia, Vue Router |
svelte |
Runes, stores, SvelteKit |
swiftui |
Views, State, Navigation, Animation |
react-native |
Components, Navigation, Lists |
flutter |
Widgets, State, Layout, Theming |
Example Workflow
User request: "Làm landing page cho dịch vụ chăm sóc da chuyên nghiệp"
AI should:
# 1. Search product type
python .claude/skills/ui-ux-pro-max/scripts/search.py "beauty spa wellness service" --domain product
# 2. Search style (based on industry: beauty, elegant)
python .claude/skills/ui-ux-pro-max/scripts/search.py "elegant minimal soft" --domain style
# 3. Search typography
python .claude/skills/ui-ux-pro-max/scripts/search.py "elegant luxury" --domain typography
# 4. Search color palette
python .claude/skills/ui-ux-pro-max/scripts/search.py "beauty spa wellness" --domain color
# 5. Search landing page structure
python .claude/skills/ui-ux-pro-max/scripts/search.py "hero-centric social-proof" --domain landing
# 6. Search UX guidelines
python .claude/skills/ui-ux-pro-max/scripts/search.py "animation" --domain ux
python .claude/skills/ui-ux-pro-max/scripts/search.py "accessibility" --domain ux
# 7. Search stack guidelines (default: html-tailwind)
python .claude/skills/ui-ux-pro-max/scripts/search.py "layout responsive" --stack html-tailwind
Then: Synthesize all search results and implement the design.
Tips for Better Results
- Be specific with keywords - "healthcare SaaS dashboard" > "app"
- Search multiple times - Different keywords reveal different insights
- Combine domains - Style + Typography + Color = Complete design system
- Always check UX - Search "animation", "z-index", "accessibility" for common issues
- Use stack flag - Get implementation-specific best practices
- Iterate - If first search doesn't match, try different keywords
Curation Constraints (UI/UX Design Principles)
The databases RECOMMEND; the 40 UI/UX Design Principles (UI-1.1–UI-9.4, the SYNC:ui-ux-design-principles block below) CONSTRAIN what you may recommend. A search hit is a CANDIDATE, not a verdict — verify it against the clauses before handing it to the user or to design. Project design-system docs outrank both the clauses and these databases; a genuine conflict goes to the user, NEVER resolved silently.
| Curated output |
Clause constraint to satisfy BEFORE recommending it |
Palette (--domain color, style palettes) |
Measure the pairs you propose — text on background ≥4.5:1, UI edges/borders ≥3:1 (UI-3.1) — and report the measured ratio; never "looks fine". Keep ONE accent with one job (UI-3.2); never let colour carry status meaning alone — pair it with an icon or label (UI-3.3). A dark variant lifts surfaces for elevation and softens pure-white text; it is NOT the light palette inverted (UI-3.4). |
Font pairing (--domain typography) |
At most 2 families × 3 weights (UI-2.1) — a pairing needing a third family is over-curated. Body renders 16px web / 17px mobile and NEVER below 14px for readable content (UI-2.2); measure 45–75 characters (UI-2.3); leading 1.5 body, 1.1–1.2 display (UI-2.4). Deliver the pairing AS a 6-step named type scale, not a loose list of sizes (UI-2.5). |
Style / product recommendation (--domain style, --domain product) |
The style must still permit ONE focal point per screen (UI-1.1) and grouping by whitespace rather than nested boxes (UI-1.3). Reject or adapt any aesthetic that requires removing the focus ring (UI-5.5), hiding labels behind placeholders (UI-7.2), or motion outside 150–250ms ease-out that ignores reduced-motion (UI-5.4). |
Spacing / layout guidance (--domain style, stack guidelines) |
Express spacing as ONE unit multiplied — 4px or 8px base, every gap a multiple (UI-4.1) — owned by the container via a gap property (UI-4.2), tighter inside than between groups (UI-4.3); breakpoints follow the content, not device names (UI-4.4). |
| Any UI handed off or built from this intelligence |
Empty, loading and error states specified FIRST, before the populated state (UI-1.5), with space reserved so loaded content never shifts the layout (UI-9.3); all 5 interaction states enumerated per interactive element — default, hover, focus, active, disabled — plus loading where it applies (UI-5.2); on mobile/touch surfaces hit targets ≥44×44pt and 8px apart (UI-8.1) with primary actions in the bottom third (UI-8.2). |
When a curated recommendation cannot satisfy a clause, say so explicitly with the clause ID and offer the compliant alternative — a beautiful palette that fails UI-3.1 is a FINDING, not a deliverable.
Common Rules for Professional UI
These are frequently overlooked issues that make UI look unprofessional:
Icons & Visual Elements
| Rule |
Do |
Don't |
| No emoji icons |
Use SVG icons (Heroicons, Lucide, Simple Icons) |
Use emojis like 🎨 🚀 ⚙️ as UI icons |
| Stable hover states |
Use color/opacity transitions on hover |
Use scale transforms that shift layout |
| Correct brand logos |
Research official SVG from Simple Icons |
Guess or use incorrect logo paths |
| Consistent icon sizing |
Use fixed viewBox (24x24) with w-6 h-6 |
Mix different icon sizes randomly |
Interaction & Cursor
| Rule |
Do |
Don't |
| Cursor pointer |
Add cursor-pointer to all clickable/hoverable cards |
Leave default cursor on interactive elements |
| Hover feedback |
Provide visual feedback (color, shadow, border) |
No indication element is interactive |
| Smooth transitions |
Use transition-colors duration-200 |
Instant state changes or too slow (>500ms) |
Light/Dark Mode Contrast
| Rule |
Do |
Don't |
| Glass card light mode |
Use bg-white/80 or higher opacity |
Use bg-white/10 (too transparent) |
| Text contrast light |
Use #0F172A (slate-900) for text |
Use #94A3B8 (slate-400) for body text |
| Muted text light |
Use #475569 (slate-600) minimum |
Use gray-400 or lighter |
| Border visibility |
Use border-gray-200 in light mode |
Use border-white/10 (invisible) |
Layout & Spacing
| Rule |
Do |
Don't |
| Floating navbar |
Add top-4 left-4 right-4 spacing |
Stick navbar to top-0 left-0 right-0 |
| Content padding |
Account for fixed navbar height |
Let content hide behind fixed elements |
| Consistent max-width |
Use same max-w-6xl or max-w-7xl |
Mix different container widths |
Async UI States & Feedback (the most-missed fundamentals)
Every screen or component that fetches, submits, or mutates data MUST render all of its states — not just the happy path. A UI that looks perfect when populated but shows a frozen blank while loading, fails silently on error, or shows nothing when empty is broken UX. Canonical state vocabulary (shared with design-spec): Default / Loading / Disabled / Error / Empty / Success.
| State |
Do |
Don't |
| Loading |
Show a spinner / skeleton / progress while any request is in-flight |
Leave a frozen or blank screen; block the whole page silently |
| Error |
Show a human-readable message with a retry / recovery action |
Fail silently, swallow the error, or dump a raw stack trace / JSON |
| Empty |
Show a meaningful empty state (illustration/message + optional CTA) |
Render a blank area when a list/table has zero items |
| In-flight / disabled |
Disable the submit/action button (or show it busy) while its request runs |
Allow double-submit by leaving the trigger active |
| Success |
Confirm completion (toast / inline / visibly updated data) |
Complete a mutation with no acknowledgment to the user |
Responsive & Multi-Device (usable on small screens)
| Rule |
Do |
Don't |
| Rows reflow |
flex-wrap: wrap or switch flex-direction: row → column at small widths |
Fixed flex-direction: row that overflows on a phone |
| Grids collapse |
Multi-column grids collapse to one column on small screens |
Fixed column counts / track widths that never collapse |
| Scroll fallback OK |
Can't reflow (table/canvas/wide grid)? Fixed min-width/min-height + overflow: auto — scrolling is fine |
Content clipped, cut off, or a control unreachable off-screen with no scroll path |
| Fluid over fixed |
min/max-width, flex-grow, %/rem sizing |
Large fixed px widths on containers/cards/dialogs |
| Big refactor → confirm |
Small-screen support needs a large layout rewrite → flag it, confirm scope with the user |
Silently rewriting a big layout, or shipping it broken on small screens |
Pre-Delivery Checklist
Before delivering UI code, verify these items:
Visual Quality
Interaction
States & Feedback (async surfaces — verify every one)
Light/Dark Mode
Layout & Responsiveness
Accessibility
Related
design --lane=marketing — Marketing/creative UI (landing pages, campaigns)
design --lane=product — Product UIs (dashboards, admin panels, SaaS apps)
shadcn-tailwind
design-spec
[IMPORTANT] Use task tracking to break ALL work into small tasks BEFORE starting — including tasks for each file read. This prevents context loss from long files. For simple tasks, AI MUST ATTENTION ask user whether to skip.
UI/UX Design Principles (Rev 1.0 — 40 clauses, web + mobile) — the working rule set for ANY task that designs, plans, implements, or reviews a user interface. Applies to BOTH platforms unless a clause names one (§8 is mobile/touch). Cite clauses by ID: UI-3.1, UI-8.2.
Precedence: project design-system / SCSS / frontend-pattern docs OUTRANK these clauses; these clauses outrank generic taste. A genuine conflict is SURFACED to the user with both sides — NEVER resolved silently. — why: the project's own recorded decision is the authority; these clauses are the default when it is silent.
1.0 Visual Hierarchy & Layout
UI-1.1 One focal point per screen. Two elements competing for first read → demote one.
UI-1.2 Signal order: size → weight → colour → position. Use the cheapest signal that works before adding another.
UI-1.3 Group by proximity, NEVER by border. Whitespace separates cleanly; boxes inside boxes do not.
UI-1.4 Align to a shared edge. Every unexplained indent reads as an accident.
UI-1.5 Design the empty, loading and error state FIRST. The full state is the easy one.
2.0 Typography
UI-2.1 Max 2 families, 3 weights each. More variety reads as inconsistency, not range.
UI-2.2 Body text 16px web, 17px mobile. NEVER below 14px for anything a user must read.
UI-2.3 Line length 45–75 characters. Constrain the measure, not the container.
UI-2.4 Leading scales inversely with size: 1.5 body, 1.1–1.2 display.
UI-2.5 Fixed type scale — 6 named steps shared with engineering. NEVER one-off sizes.
3.0 Colour & Contrast
UI-3.1 Contrast 4.5:1 text, 3:1 UI edges. Measure it — NEVER judge by eye on a bright screen.
UI-3.2 One accent, one job. An accent that is everywhere points at nothing.
UI-3.3 Colour NEVER carries meaning alone — pair it with an icon, label or position.
UI-3.4 Dark mode is NOT inverted light mode. Lift surfaces to signal elevation; soften pure-white text.
4.0 Spacing & Grid
UI-4.1 One spacing unit, multiplied — 4px or 8px base; every gap a multiple of it.
UI-4.2 Space belongs to the container, not the child. Use gap; reserve margins for exceptions.
UI-4.3 Tighter inside, looser between — inner padding always smaller than the gap to the next group.
UI-4.4 Breakpoints follow content, not devices. Break where the layout stops working.
5.0 Interaction & Feedback
UI-5.1 Every action gets a response under 100ms, even when the result takes longer.
UI-5.2 Specify all 5 states — default, hover, focus, active, disabled — plus loading where it applies.
UI-5.3 Prefer undo over confirmation. Confirm ONLY what cannot be reversed.
UI-5.4 Motion clarifies cause and effect: 150–250ms, ease-out, honours reduced-motion.
UI-5.5 Keep the visible focus ring. Restyle it if it clashes; NEVER remove it.
6.0 Navigation & IA
UI-6.1 Every screen answers: where am I, what's here, where next.
UI-6.2 Max 5 top-level destinations. Depth beats a crowded first level.
UI-6.3 Label by the user's word, not the internal one. Team vocabulary is not a taxonomy.
UI-6.4 Every state deserves a URL or a back path. Deep links and hardware back must land somewhere sensible.
7.0 Forms & Input
UI-7.1 Ask for less. Every field needs a reason it exists today.
UI-7.2 Labels stay visible. Placeholders are hints, NEVER labels.
UI-7.3 Validate on blur, not on keystroke. Errors sit next to the field and say how to fix it.
UI-7.4 Match keyboard to data type — correct input type, autocomplete and autocapitalise on every field.
UI-7.5 NEVER lose entered data. Preserve input across errors, navigation and refresh.
8.0 Mobile & Touch (mobile/touch surfaces)
UI-8.1 Hit target ≥44×44pt, 8px apart. The target may exceed the visible icon.
UI-8.2 Primary actions in the bottom third — that is where the thumb lives.
UI-8.3 Gestures are shortcuts, NEVER the only route. Anything swipeable is also tappable.
UI-8.4 Respect safe areas and the keyboard. Notch, home indicator and on-screen keyboard all steal space.
9.0 Speed & Perceived Speed
UI-9.1 Show structure before data — skeletons for known layouts, spinners only for unknown waits.
UI-9.2 Assume success optimistically. Update UI first, reconcile after, roll back visibly on failure.
UI-9.3 Reserve space for anything that loads. Images, ads and fonts must NEVER shift the layout.
UI-9.4 Design for the slow connection. Offline, timeout and retry are states, NOT edge cases.
Apply by role — the clauses are one set; what you DO with them depends on the task:
| Role |
Obligation |
| DESIGN / PLAN a surface |
Clauses shape the artifact: empty/loading/error specified first (UI-1.5), all 5 states enumerated (UI-5.2), type scale + spacing unit declared (UI-2.5, UI-4.1) |
| IMPLEMENT a component |
Pre-completion gate: states · tokens · contrast · focus ring · touch target · reserved space (UI-5.2, UI-2.5/UI-4.1, UI-3.1, UI-5.5, UI-8.1, UI-9.3) |
| REVIEW UI code or a design artifact |
Each clause is a fail-condition; every finding cites UI-<clause> + file:line + severity. NEVER a tick-box sweep — one focused pass per section |
| SCAN / document a UI system |
Record where the PROJECT deliberately deviates, so the overriding doc becomes the recorded authority |
Skip ONLY when the change has no user-facing surface (backend-only, tooling, docs) — state that explicitly so the skip is auditable, not an omission.
AI Mistake Prevention — Failure modes to avoid on every task:
Re-read files after context changes. Context compaction, resume, or long-running work can make memory stale; verify current files before acting.
Verify generated content against source evidence. AI hallucinates APIs, names, claims, and document facts. Check the relevant source before documenting or referencing.
Check downstream references before deleting or renaming. Removing an artifact can stale docs, generated mirrors, configs, and callers; map references first.
Trace the full impact chain after edits. Changing a definition can miss derived outputs and consumers. Follow the affected chain before declaring done.
Verify ALL affected outputs, not just the first. One green check is not all green checks; validate every output surface the change can affect.
Assume existing values are intentional — ask WHY before changing OR flagging one as a defect. Before changing or reporting a constant, limit, flag, cutoff, wording, or pattern, read nearby context and history, the CALLER's ordering, and 2+ sibling call sites of the same convention. A doc stating WHAT without WHY is missing rationale, not proof of a missing guard.
Surface ambiguity before acting — don't pick silently. Multiple valid interpretations require an explicit question or stated assumption with risk.
Assert the outcome your system owns, not the intermediate state your infrastructure owns. When verifying async work, assert the final business state — never the delivery/retry bookkeeping held in shared infrastructure that any co-running process can write. Such a check passes when run alone and flakes the moment anything else shares that infrastructure.
Keep shared guidance role-relevant. Universal guidance must help every receiving skill or agent; code-specific obligations belong only in code-specific protocols.
Critical Thinking Mindset — Apply critical thinking, sequential thinking. Every claim needs traced proof, confidence >80% to act.
Anti-hallucination: Never present guess as fact — cite sources for every claim, admit uncertainty freely, self-check output for errors, cross-reference independently, stay skeptical of own confidence — certainty without evidence root of all hallucination.
MUST ATTENTION apply critical + sequential thinking — every claim needs appropriate traced evidence (file:line for repo/code claims; source URL or artifact section for research, product, content, and docs claims); confidence >80% to act, <60% DO NOT recommend. Anti-hallucination: never present guess as fact, admit uncertainty freely, cross-reference independently, stay skeptical of own confidence.
MUST ATTENTION apply AI mistake prevention — verify generated content against evidence, trace downstream references before deleting or renaming, verify all affected outputs, re-read files after context loss, and surface ambiguity before acting.
- MUST ATTENTION apply the 40 UI/UX Design Principles (
UI-1.1–UI-9.4) to any user-facing surface: one focal point, proximity grouping, empty/loading/error designed FIRST (§1) · ≤2 families, 16px web / 17px mobile body, never <14px, 45–75ch, fixed 6-step scale (§2) · 4.5:1 text / 3:1 edges measured, one accent, colour never alone, dark mode ≠ inversion (§3) · one 4/8px unit, gap over margins, tighter-inside-looser-between, content-driven breakpoints (§4) · <100ms response, all 5 states, undo over confirm, 150–250ms ease-out honouring reduced-motion, visible focus ring (§5) · where-am-I/what's-here/where-next, ≤5 top-level destinations, user's words, URL or back path (§6) · fewer fields, visible labels, validate on blur, matched keyboard, NEVER lose input (§7) · ≥44×44pt targets 8px apart, bottom-third primaries, gestures never the only route, safe areas (§8) · structure before data, optimistic update with visible rollback, reserved space, slow-connection states (§9). Project design-system docs OUTRANK these clauses — a genuine conflict goes to the user, NEVER resolved silently. Cite every finding as UI-<clause> + file:line. Skip ONLY for changes with no user-facing surface, stated explicitly.
Project Protocol Overlay — Before executing this skill, resolve any PROJECT overlay rules layered onto it: match this skill's name against the Target column of the project's skill-protocol index (docs/project-reference/skill-protocols-reference.md by default; a referenceDocs entry in docs/project-config.json overrides the path), taking the most specific matching tier ONLY — exact name > glob > *. That precedence orders overlays against EACH OTHER, never against this skill. Read ONLY the matched bodies, resolved as <protocols-dir>/<Name>.md; a row's Body link is display text, never a read path. A matched body that is missing or malformed is REPORTED and skipped — never reconstructed from the index Description. No index, or no match -> proceed with no overlay, silently. Full contract: .claude/skills/project-skill-protocol/references/registry.md.
Overlays are ADDITIVE ONLY: they ADD rules on top of this skill's own protocol and NEVER replace, override, disable, or reinterpret a rule it already states — removing every overlay must return this skill to exactly its documented behavior. An overlay is a BRIEF, not an authority escalation: it can NEVER waive a workflow gate, git discipline, a review gate, or a user-confirmation gate. A genuine overlay-vs-skill conflict, or two equally-specific overlays that directly contradict -> surface both to the user; NEVER resolve silently.
MUST ATTENTION resolve project protocol overlays for this skill BEFORE executing — most specific matching tier only (exact > glob > *, which ranks overlays against each other, NEVER against this skill), read only matched bodies at <protocols-dir>/<Name>.md; a missing or malformed body is reported, never reconstructed. Overlays are ADDITIVE ONLY (they never replace this skill's own rules) and are a brief, NEVER an authority escalation; an equal-specificity contradiction goes to the user.
Closing Reminders
IMPORTANT MUST ATTENTION Goal: Provide searchable UI/UX design intelligence — 50 styles, 21 palettes, 50 font pairings, 20 charts, 8 stacks — synthesized into accessible, professional, contrast-correct interfaces.
Protocols in force (concise digest of the SYNC/shared blocks this skill carries):
- AI Mistake Prevention: verify generated content against evidence, trace downstream references, verify all affected outputs, re-read after context loss, surface ambiguity.
- Critical Thinking: Traced
file:line proof per claim, confidence >80% to act, never guess.
IMPORTANT MUST ATTENTION break work into small todo tasks using task tracking BEFORE starting
IMPORTANT MUST ATTENTION search codebase for 3+ similar patterns before creating new code
IMPORTANT MUST ATTENTION cite file:line evidence for every claim (confidence >80% to act)
IMPORTANT MUST ATTENTION add a final review todo task to verify work quality
IMPORTANT MUST ATTENTION the 40 UI/UX Design Principles (UI-1.1–UI-9.4) CONSTRAIN the curation — a curated palette must still measure 4.5:1 text / 3:1 edges (UI-3.1) with one accent (UI-3.2) and colour never alone (UI-3.3); a curated font pairing must still be ≤2 families × 3 weights (UI-2.1) at 16px web / 17px mobile, never <14px (UI-2.2), delivered as a 6-step type scale (UI-2.5); spacing is ONE 4/8px unit multiplied (UI-4.1); anything built from it specifies empty/loading/error FIRST (UI-1.5) and all 5 interaction states (UI-5.2) — a search hit that fails a clause is a finding, NOT a deliverable
[TASK-PLANNING] Before acting, analyze task scope and systematically break it into small todo tasks and sub-tasks using task tracking.
Hookless Prompt Protocol Mirror (Auto-Synced)
Source: .claude/.ck.json + .claude/skills/shared/sync-inline-versions.md (:full blocks) + .claude/scripts/lib/hookless-prompt-protocol.cjs
[WORKFLOW-EXECUTION-PROTOCOL] [BLOCKING] Workflow Execution Protocol — MANDATORY IMPORTANT MUST CRITICAL. Do not skip for any reason.
Generic portability boundary: Reusable skills and protocol text stay project-neutral; project-specific conventions are discovered from docs/project-config.json and docs/project-reference/. Apply shared AI-SDD from shared/sdd-artifact-contract.md. Read docs/project-config.json and docs/project-reference/docs-index-reference.md, then open the project reference docs named there. For spec, test-case, behavior-change, public-contract, or docs/specs/ work, route through the local spec docs named by the docs index: feature-spec-reference.md, spec-system-reference.md, spec-principles.md, and workflow-spec-test-code-cycle-reference.md when specs/tests/code must stay synchronized. If either file or a required reference d
…(truncated)
1---
2name: ui-ux-pro-max
3description: '[Frontend] Use when designing beautiful or aesthetic interfaces with curated UI styles, palettes, fonts, charts, and stacks.'
4metadata:
5 source: https://skillmd.com/skills/duc01226/ui-ux-pro-max
6---
7
8> Codex compatibility note:
9>
10> - Invoke repository skills with `$skill-name` in Codex; this mirrored copy rewrites legacy Claude `/skill-name` references.
11> - Task tracker mandate: BEFORE executing any workflow or skill step, create/update task tracking for all steps and keep it synchronized as progress changes.
12> - User-question prompts mean to ask the user directly in Codex.
13> - Ignore Claude-specific mode-switch instructions when they appear.
14> - Strict execution contract: when a user explicitly invokes a skill, execute that skill protocol as written.
15> - Subagent authorization: when a skill is user-invoked or AI-detected and its protocol requires subagents, that skill activation authorizes use of the required `spawn_agent` subagent(s) for that task.
16> - Do not skip, reorder, or merge protocol steps unless the user explicitly approves the deviation first.
17> - For workflow skills, execute each listed child-skill step explicitly and report step-by-step evidence.
18> - If a required step/tool cannot run in this environment, stop and ask the user before adapting.
19
20<!-- CODEX:PROJECT-REFERENCE-LOADING:START -->
21
22## Codex Project-Reference Loading (No Hooks)
23
24Codex uses static project-reference loading instead of runtime-injected project docs.
25When coding, planning, debugging, testing, or reviewing, open project docs explicitly using this routing.
26
27**Always read:**
28
29- `docs/project-config.json` (project-specific paths, commands, modules, and workflow/test settings)
30- `docs/project-reference/docs-index-reference.md` (routes to the full `docs/project-reference/*` catalog)
31- `docs/project-reference/lessons.md` (always-on guardrails and anti-patterns)
32
33**Missing/stale context route:** If `docs/project-config.json`, the docs index, `lessons.md`, `CLAUDE.md`, `AGENTS.md`, or any task-required reference doc is missing or stale, auto-run `$project-init` or the narrow setup route (`$project-config`, `$docs-init`, `$scan-all`, `$scan --target=<key>`, `$claude-md-init`) before ordinary project-specific work. If Codex mirrors or `AGENTS.md` are missing/stale, ask the user to run `$sync-codex`; do not auto-run it.
34
35**Situation-based docs:**
36
37- Project structure/architecture/tech-stack/deployment/setup (any layer — backend, frontend, or infra): `project-structure-reference.md`
38- Backend/CQRS/API/domain/entity changes: `backend-patterns-reference.md`, `domain-entities-reference.md`
39- Frontend/UI/styling/design-system: `frontend-patterns-reference.md`, `scss-styling-guide.md`, `design-system/README.md`
40- Spec authoring, `docs/specs/` pathing, or TC format: `feature-spec-reference.md`, `spec-system-reference.md`, `spec-principles.md`
41- Behavior/public-contract changes or spec-test-code sync: `workflow-spec-test-code-cycle-reference.md` plus the spec docs above
42- Derived spec indexes/ERDs/reimplementation guides: `spec-system-reference.md` and source Feature Specs under `docs/specs/`
43- Integration test implementation/review: `integration-test-reference.md`
44- E2E test implementation/review: `e2e-test-reference.md`
45- Code review/audit work: `code-review-rules.md` plus domain docs above based on changed files
46
47Do not read all docs blindly. Start from `docs-index-reference.md`, then open only relevant files for the task.
48
49<!-- CODEX:PROJECT-REFERENCE-LOADING:END -->
50
51## Quick Summary
52
53**Goal:** Provide searchable UI/UX design intelligence -- 50 styles, 21 palettes, 50 font pairings, 20 charts, 8 stacks.
54
55**Workflow:**
56
571. **Analyze** — Extract product type, style keywords, industry, and stack from user request
582. **Search** — Query `search.py` across domains: product, style, typography, color, landing, chart, UX
593. **Stack Guidelines** — Get stack-specific best practices (default: html-tailwind)
604. **Implement** — Synthesize search results into cohesive design and code
61
62**Key Rules:**
63
64- No emojis as UI icons (use SVG: Heroicons, Lucide, Simple Icons)
65- All clickable elements need `cursor-pointer` and hover feedback
66- Light mode text must have 4.5:1 minimum contrast ratio
67- **Every async surface MUST handle all its states** — show a loading indicator while in-flight, a user-visible error (with retry) on failure, and a meaningful empty state when there's no data. Never a frozen blank, a silent failure, or a blank list.
68- **Responsive by default** — layouts MUST stay usable on small devices. Preferred: reflow (rows `flex-wrap` or `row → column`, grids collapse to one column). Acceptable fallback when a layout genuinely can't reflow (tables, canvases, wide grids): fixed `min-width`/`min-height` + `overflow: auto` scroll — scrolling is OK. Hard minimum: nothing broken — no clipped, cut-off, or unreachable content. If small-screen support needs a refactor too big for the task, confirm scope with the user first.
69- Test both light/dark modes, all UI states (loading/error/empty), and responsive breakpoints (320/768/1024/1440) before delivery
70
71**Pre-read (project design system):** When the project has a design system, load `designSystem.canonicalDoc` + `tokenFiles` from `docs/project-config.json` and prefer project tokens/components over the generic style intelligence database.
72
73**Be skeptical. Apply critical thinking, sequential thinking. Every claim needs traced proof, confidence percentages (Idea should be more than 80%).**
74
75# UI/UX Pro Max - Design Intelligence
76
77Searchable database of UI styles, color palettes, font pairings, chart types, product recommendations, UX guidelines, and stack-specific best practices.
78
79## Prerequisites
80
81Check if Python is installed:
82
83```bash
84python3 --version || python --version
85```
86
87If Python is not installed, install it based on user's OS:
88
89**macOS:**
90
91```bash
92brew install python3
93```
94
95**Ubuntu/Debian:**
96
97```bash
98sudo apt update && sudo apt install python3
99```
100
101**Windows:**
102
103```powershell
104winget install Python.Python.3.12
105```
106
107---
108
109## How to Use This Skill
110
111When user requests UI/UX work (design, build, create, implement, review, fix, improve), follow this workflow:
112
113### Step 1: Analyze User Requirements
114
115Extract key information from user request:
116
117- **Product type**: SaaS, e-commerce, portfolio, dashboard, landing page, etc.
118- **Style keywords**: minimal, playful, professional, elegant, dark mode, etc.
119- **Industry**: healthcare, fintech, gaming, education, etc.
120- **Stack**: React, Vue, Next.js, or default to `html-tailwind`
121
122### Step 2: Search Relevant Domains
123
124Use `search.py` multiple times to gather comprehensive information. Search until you have enough context.
125
126```bash
127python .claude/skills/ui-ux-pro-max/scripts/search.py "<keyword>" --domain <domain> [-n <max_results>]
128```
129
130**Recommended search order:**
131
1321. **Product** - Get style recommendations for product type
1332. **Style** - Get detailed style guide (colors, effects, frameworks)
1343. **Typography** - Get font pairings with Google Fonts imports
1354. **Color** - Get color palette (Primary, Secondary, CTA, Background, Text, Border)
1365. **Landing** - Get page structure (if landing page)
1376. **Chart** - Get chart recommendations (if dashboard/analytics)
1387. **UX** - Get best practices and anti-patterns
1398. **Stack** - Get stack-specific guidelines (default: html-tailwind)
140
141### Step 3: Stack Guidelines (Default: html-tailwind)
142
143If user doesn't specify a stack, **default to `html-tailwind`**.
144
145```bash
146python .claude/skills/ui-ux-pro-max/scripts/search.py "<keyword>" --stack html-tailwind
147```
148
149Available stacks: `html-tailwind`, `react`, `nextjs`, `vue`, `svelte`, `swiftui`, `react-native`, `flutter`
150
151---
152
153## Search Reference
154
155### Available Domains
156
157| Domain | Use For | Example Keywords |
158| ------------ | ------------------------------------ | -------------------------------------------------------- |
159| `product` | Product type recommendations | SaaS, e-commerce, portfolio, healthcare, beauty, service |
160| `style` | UI styles, colors, effects | glassmorphism, minimalism, dark mode, brutalism |
161| `typography` | Font pairings, Google Fonts | elegant, playful, professional, modern |
162| `color` | Color palettes by product type | saas, ecommerce, healthcare, beauty, fintech, service |
163| `landing` | Page structure, CTA strategies | hero, hero-centric, testimonial, pricing, social-proof |
164| `chart` | Chart types, library recommendations | trend, comparison, timeline, funnel, pie |
165| `ux` | Best practices, anti-patterns | animation, accessibility, z-index, loading |
166| `prompt` | AI prompts, CSS keywords | (style name) |
167
168### Available Stacks
169
170| Stack | Focus |
171| --------------- | ---------------------------------------------- |
172| `html-tailwind` | Tailwind utilities, responsive, a11y (DEFAULT) |
173| `react` | State, hooks, performance, patterns |
174| `nextjs` | SSR, routing, images, API routes |
175| `vue` | Composition API, Pinia, Vue Router |
176| `svelte` | Runes, stores, SvelteKit |
177| `swiftui` | Views, State, Navigation, Animation |
178| `react-native` | Components, Navigation, Lists |
179| `flutter` | Widgets, State, Layout, Theming |
180
181---
182
183## Example Workflow
184
185**User request:** "Làm landing page cho dịch vụ chăm sóc da chuyên nghiệp"
186
187**AI should:**
188
189```bash
190# 1. Search product type
191python .claude/skills/ui-ux-pro-max/scripts/search.py "beauty spa wellness service" --domain product
192
193# 2. Search style (based on industry: beauty, elegant)
194python .claude/skills/ui-ux-pro-max/scripts/search.py "elegant minimal soft" --domain style
195
196# 3. Search typography
197python .claude/skills/ui-ux-pro-max/scripts/search.py "elegant luxury" --domain typography
198
199# 4. Search color palette
200python .claude/skills/ui-ux-pro-max/scripts/search.py "beauty spa wellness" --domain color
201
202# 5. Search landing page structure
203python .claude/skills/ui-ux-pro-max/scripts/search.py "hero-centric social-proof" --domain landing
204
205# 6. Search UX guidelines
206python .claude/skills/ui-ux-pro-max/scripts/search.py "animation" --domain ux
207python .claude/skills/ui-ux-pro-max/scripts/search.py "accessibility" --domain ux
208
209# 7. Search stack guidelines (default: html-tailwind)
210python .claude/skills/ui-ux-pro-max/scripts/search.py "layout responsive" --stack html-tailwind
211```
212
213**Then:** Synthesize all search results and implement the design.
214
215---
216
217## Tips for Better Results
218
2191. **Be specific with keywords** - "healthcare SaaS dashboard" > "app"
2202. **Search multiple times** - Different keywords reveal different insights
2213. **Combine domains** - Style + Typography + Color = Complete design system
2224. **Always check UX** - Search "animation", "z-index", "accessibility" for common issues
2235. **Use stack flag** - Get implementation-specific best practices
2246. **Iterate** - If first search doesn't match, try different keywords
225
226---
227
228## Curation Constraints (UI/UX Design Principles)
229
230The databases RECOMMEND; the 40 UI/UX Design Principles (`UI-1.1`–`UI-9.4`, the `SYNC:ui-ux-design-principles` block below) CONSTRAIN what you may recommend. A search hit is a CANDIDATE, not a verdict — verify it against the clauses before handing it to the user or to `design`. Project design-system docs outrank both the clauses and these databases; a genuine conflict goes to the user, NEVER resolved silently.
231
232| Curated output | Clause constraint to satisfy BEFORE recommending it |
233| ------------------------------------------------------------------------- | ----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- |
234| **Palette** (`--domain color`, style palettes) | Measure the pairs you propose — text on background ≥4.5:1, UI edges/borders ≥3:1 (`UI-3.1`) — and report the measured ratio; never "looks fine". Keep ONE accent with one job (`UI-3.2`); never let colour carry status meaning alone — pair it with an icon or label (`UI-3.3`). A dark variant lifts surfaces for elevation and softens pure-white text; it is NOT the light palette inverted (`UI-3.4`). |
235| **Font pairing** (`--domain typography`) | At most 2 families × 3 weights (`UI-2.1`) — a pairing needing a third family is over-curated. Body renders 16px web / 17px mobile and NEVER below 14px for readable content (`UI-2.2`); measure 45–75 characters (`UI-2.3`); leading 1.5 body, 1.1–1.2 display (`UI-2.4`). Deliver the pairing AS a 6-step named type scale, not a loose list of sizes (`UI-2.5`). |
236| **Style / product recommendation** (`--domain style`, `--domain product`) | The style must still permit ONE focal point per screen (`UI-1.1`) and grouping by whitespace rather than nested boxes (`UI-1.3`). Reject or adapt any aesthetic that requires removing the focus ring (`UI-5.5`), hiding labels behind placeholders (`UI-7.2`), or motion outside 150–250ms ease-out that ignores reduced-motion (`UI-5.4`). |
237| **Spacing / layout guidance** (`--domain style`, stack guidelines) | Express spacing as ONE unit multiplied — 4px or 8px base, every gap a multiple (`UI-4.1`) — owned by the container via a gap property (`UI-4.2`), tighter inside than between groups (`UI-4.3`); breakpoints follow the content, not device names (`UI-4.4`). |
238| **Any UI handed off or built from this intelligence** | Empty, loading and error states specified FIRST, before the populated state (`UI-1.5`), with space reserved so loaded content never shifts the layout (`UI-9.3`); all 5 interaction states enumerated per interactive element — default, hover, focus, active, disabled — plus loading where it applies (`UI-5.2`); on mobile/touch surfaces hit targets ≥44×44pt and 8px apart (`UI-8.1`) with primary actions in the bottom third (`UI-8.2`). |
239
240**When a curated recommendation cannot satisfy a clause,** say so explicitly with the clause ID and offer the compliant alternative — a beautiful palette that fails `UI-3.1` is a FINDING, not a deliverable.
241
242## Common Rules for Professional UI
243
244These are frequently overlooked issues that make UI look unprofessional:
245
246### Icons & Visual Elements
247
248| Rule | Do | Don't |
249| -------------------------- | ----------------------------------------------- | -------------------------------------- |
250| **No emoji icons** | Use SVG icons (Heroicons, Lucide, Simple Icons) | Use emojis like 🎨 🚀 ⚙️ as UI icons |
251| **Stable hover states** | Use color/opacity transitions on hover | Use scale transforms that shift layout |
252| **Correct brand logos** | Research official SVG from Simple Icons | Guess or use incorrect logo paths |
253| **Consistent icon sizing** | Use fixed viewBox (24x24) with w-6 h-6 | Mix different icon sizes randomly |
254
255### Interaction & Cursor
256
257| Rule | Do | Don't |
258| ---------------------- | ----------------------------------------------------- | -------------------------------------------- |
259| **Cursor pointer** | Add `cursor-pointer` to all clickable/hoverable cards | Leave default cursor on interactive elements |
260| **Hover feedback** | Provide visual feedback (color, shadow, border) | No indication element is interactive |
261| **Smooth transitions** | Use `transition-colors duration-200` | Instant state changes or too slow (>500ms) |
262
263### Light/Dark Mode Contrast
264
265| Rule | Do | Don't |
266| ------------------------- | ----------------------------------- | --------------------------------------- |
267| **Glass card light mode** | Use `bg-white/80` or higher opacity | Use `bg-white/10` (too transparent) |
268| **Text contrast light** | Use `#0F172A` (slate-900) for text | Use `#94A3B8` (slate-400) for body text |
269| **Muted text light** | Use `#475569` (slate-600) minimum | Use gray-400 or lighter |
270| **Border visibility** | Use `border-gray-200` in light mode | Use `border-white/10` (invisible) |
271
272### Layout & Spacing
273
274| Rule | Do | Don't |
275| ------------------------ | ----------------------------------- | -------------------------------------- |
276| **Floating navbar** | Add `top-4 left-4 right-4` spacing | Stick navbar to `top-0 left-0 right-0` |
277| **Content padding** | Account for fixed navbar height | Let content hide behind fixed elements |
278| **Consistent max-width** | Use same `max-w-6xl` or `max-w-7xl` | Mix different container widths |
279
280### Async UI States & Feedback (the most-missed fundamentals)
281
282> **Every screen or component that fetches, submits, or mutates data MUST render all of its states — not just the happy path.** A UI that looks perfect when populated but shows a frozen blank while loading, fails silently on error, or shows nothing when empty is broken UX. Canonical state vocabulary (shared with `design-spec`): **Default / Loading / Disabled / Error / Empty / Success**.
283
284| State | Do | Don't |
285| ------------------------ | ------------------------------------------------------------------------- | ------------------------------------------------------------------ |
286| **Loading** | Show a spinner / skeleton / progress while any request is in-flight | Leave a frozen or blank screen; block the whole page silently |
287| **Error** | Show a human-readable message with a retry / recovery action | Fail silently, swallow the error, or dump a raw stack trace / JSON |
288| **Empty** | Show a meaningful empty state (illustration/message + optional CTA) | Render a blank area when a list/table has zero items |
289| **In-flight / disabled** | Disable the submit/action button (or show it busy) while its request runs | Allow double-submit by leaving the trigger active |
290| **Success** | Confirm completion (toast / inline / visibly updated data) | Complete a mutation with no acknowledgment to the user |
291
292### Responsive & Multi-Device (usable on small screens)
293
294| Rule | Do | Don't |
295| -------------------------- | ------------------------------------------------------------------------------------------------------------ | --------------------------------------------------------------------------------- |
296| **Rows reflow** | `flex-wrap: wrap` or switch `flex-direction: row → column` at small widths | Fixed `flex-direction: row` that overflows on a phone |
297| **Grids collapse** | Multi-column grids collapse to one column on small screens | Fixed column counts / track widths that never collapse |
298| **Scroll fallback OK** | Can't reflow (table/canvas/wide grid)? Fixed `min-width`/`min-height` + `overflow: auto` — scrolling is fine | Content clipped, cut off, or a control unreachable off-screen with no scroll path |
299| **Fluid over fixed** | `min/max-width`, `flex-grow`, `%`/`rem` sizing | Large fixed px widths on containers/cards/dialogs |
300| **Big refactor → confirm** | Small-screen support needs a large layout rewrite → flag it, confirm scope with the user | Silently rewriting a big layout, or shipping it broken on small screens |
301
302---
303
304## Pre-Delivery Checklist
305
306Before delivering UI code, verify these items:
307
308### Visual Quality
309
310- [ ] No emojis used as icons (use SVG instead)
311- [ ] All icons from consistent icon set (Heroicons/Lucide)
312- [ ] Brand logos are correct (verified from Simple Icons)
313- [ ] Hover states don't cause layout shift
314
315### Interaction
316
317- [ ] All clickable elements have `cursor-pointer`
318- [ ] Hover states provide clear visual feedback
319- [ ] Transitions are smooth (150-300ms)
320- [ ] Focus states visible for keyboard navigation
321
322### States & Feedback (async surfaces — verify every one)
323
324- [ ] **Loading** — every data fetch / submit / mutation shows a spinner or skeleton while in-flight (never a frozen blank screen)
325- [ ] **Error** — every operation that can fail shows a human-readable error with a retry / recovery path (no silent failure, no raw stack trace)
326- [ ] **Empty** — every list / table / collection view has a meaningful empty state (message + optional CTA), not a blank area
327- [ ] **In-flight disable** — submit/action buttons are disabled or busy while their request runs (no double-submit)
328- [ ] **Success** — completed mutations give confirmation feedback (toast / inline / updated data)
329
330### Light/Dark Mode
331
332- [ ] Light mode text has sufficient contrast (4.5:1 minimum)
333- [ ] Glass/transparent elements visible in light mode
334- [ ] Borders visible in both modes
335- [ ] Test both modes before delivery
336
337### Layout & Responsiveness
338
339- [ ] Floating elements have proper spacing from edges
340- [ ] No content hidden behind fixed navbars
341- [ ] Responsive at 320px, 768px, 1024px, 1440px
342- [ ] Rows `flex-wrap` or switch `row → column` on small screens; multi-column grids collapse to one column
343- [ ] Usable down to 320px — **nothing broken** (no clipped, cut-off, or unreachable content); where a layout genuinely can't reflow, a `min-width`/`min-height` + `overflow: auto` scroll is acceptable
344- [ ] Any small-screen fix that needs a large layout refactor was flagged and confirmed with the user (not silently rewritten)
345
346### Accessibility
347
348- [ ] All images have alt text
349- [ ] Form inputs have labels
350- [ ] Color is not the only indicator
351- [ ] `prefers-reduced-motion` respected
352
353## Related
354
355- `design --lane=marketing` — Marketing/creative UI (landing pages, campaigns)
356- `design --lane=product` — Product UIs (dashboards, admin panels, SaaS apps)
357- `shadcn-tailwind`
358- `design-spec`
359
360---
361
362> **[IMPORTANT]** Use task tracking to break ALL work into small tasks BEFORE starting — including tasks for each file read. This prevents context loss from long files. For simple tasks, AI MUST ATTENTION ask user whether to skip.
363
364<!-- SYNC:ui-ux-design-principles -->
365
366> **UI/UX Design Principles (Rev 1.0 — 40 clauses, web + mobile)** — the working rule set for ANY task that designs, plans, implements, or reviews a user interface. Applies to BOTH platforms unless a clause names one (§8 is mobile/touch). Cite clauses by ID: `UI-3.1`, `UI-8.2`.
367>
368> **Precedence:** project design-system / SCSS / frontend-pattern docs OUTRANK these clauses; these clauses outrank generic taste. A genuine conflict is SURFACED to the user with both sides — NEVER resolved silently. — why: the project's own recorded decision is the authority; these clauses are the default when it is silent.
369>
370> **1.0 Visual Hierarchy & Layout**
371>
372> - `UI-1.1` One focal point per screen. Two elements competing for first read → demote one.
373> - `UI-1.2` Signal order: size → weight → colour → position. Use the cheapest signal that works before adding another.
374> - `UI-1.3` Group by proximity, NEVER by border. Whitespace separates cleanly; boxes inside boxes do not.
375> - `UI-1.4` Align to a shared edge. Every unexplained indent reads as an accident.
376> - `UI-1.5` Design the empty, loading and error state FIRST. The full state is the easy one.
377>
378> **2.0 Typography**
379>
380> - `UI-2.1` Max 2 families, 3 weights each. More variety reads as inconsistency, not range.
381> - `UI-2.2` Body text 16px web, 17px mobile. NEVER below 14px for anything a user must read.
382> - `UI-2.3` Line length 45–75 characters. Constrain the measure, not the container.
383> - `UI-2.4` Leading scales inversely with size: 1.5 body, 1.1–1.2 display.
384> - `UI-2.5` Fixed type scale — 6 named steps shared with engineering. NEVER one-off sizes.
385>
386> **3.0 Colour & Contrast**
387>
388> - `UI-3.1` Contrast 4.5:1 text, 3:1 UI edges. Measure it — NEVER judge by eye on a bright screen.
389> - `UI-3.2` One accent, one job. An accent that is everywhere points at nothing.
390> - `UI-3.3` Colour NEVER carries meaning alone — pair it with an icon, label or position.
391> - `UI-3.4` Dark mode is NOT inverted light mode. Lift surfaces to signal elevation; soften pure-white text.
392>
393> **4.0 Spacing & Grid**
394>
395> - `UI-4.1` One spacing unit, multiplied — 4px or 8px base; every gap a multiple of it.
396> - `UI-4.2` Space belongs to the container, not the child. Use `gap`; reserve margins for exceptions.
397> - `UI-4.3` Tighter inside, looser between — inner padding always smaller than the gap to the next group.
398> - `UI-4.4` Breakpoints follow content, not devices. Break where the layout stops working.
399>
400> **5.0 Interaction & Feedback**
401>
402> - `UI-5.1` Every action gets a response under 100ms, even when the result takes longer.
403> - `UI-5.2` Specify all 5 states — default, hover, focus, active, disabled — plus loading where it applies.
404> - `UI-5.3` Prefer undo over confirmation. Confirm ONLY what cannot be reversed.
405> - `UI-5.4` Motion clarifies cause and effect: 150–250ms, ease-out, honours reduced-motion.
406> - `UI-5.5` Keep the visible focus ring. Restyle it if it clashes; NEVER remove it.
407>
408> **6.0 Navigation & IA**
409>
410> - `UI-6.1` Every screen answers: where am I, what's here, where next.
411> - `UI-6.2` Max 5 top-level destinations. Depth beats a crowded first level.
412> - `UI-6.3` Label by the user's word, not the internal one. Team vocabulary is not a taxonomy.
413> - `UI-6.4` Every state deserves a URL or a back path. Deep links and hardware back must land somewhere sensible.
414>
415> **7.0 Forms & Input**
416>
417> - `UI-7.1` Ask for less. Every field needs a reason it exists today.
418> - `UI-7.2` Labels stay visible. Placeholders are hints, NEVER labels.
419> - `UI-7.3` Validate on blur, not on keystroke. Errors sit next to the field and say how to fix it.
420> - `UI-7.4` Match keyboard to data type — correct input type, autocomplete and autocapitalise on every field.
421> - `UI-7.5` NEVER lose entered data. Preserve input across errors, navigation and refresh.
422>
423> **8.0 Mobile & Touch** _(mobile/touch surfaces)_
424>
425> - `UI-8.1` Hit target ≥44×44pt, 8px apart. The target may exceed the visible icon.
426> - `UI-8.2` Primary actions in the bottom third — that is where the thumb lives.
427> - `UI-8.3` Gestures are shortcuts, NEVER the only route. Anything swipeable is also tappable.
428> - `UI-8.4` Respect safe areas and the keyboard. Notch, home indicator and on-screen keyboard all steal space.
429>
430> **9.0 Speed & Perceived Speed**
431>
432> - `UI-9.1` Show structure before data — skeletons for known layouts, spinners only for unknown waits.
433> - `UI-9.2` Assume success optimistically. Update UI first, reconcile after, roll back visibly on failure.
434> - `UI-9.3` Reserve space for anything that loads. Images, ads and fonts must NEVER shift the layout.
435> - `UI-9.4` Design for the slow connection. Offline, timeout and retry are states, NOT edge cases.
436>
437> **Apply by role** — the clauses are one set; what you DO with them depends on the task:
438>
439> | Role | Obligation |
440> | ----------------------------------- | ----------------------------------------------------------------------------------------------------------------------------------------------------------------------- |
441> | DESIGN / PLAN a surface | Clauses shape the artifact: empty/loading/error specified first (`UI-1.5`), all 5 states enumerated (`UI-5.2`), type scale + spacing unit declared (`UI-2.5`, `UI-4.1`) |
442> | IMPLEMENT a component | Pre-completion gate: states · tokens · contrast · focus ring · touch target · reserved space (`UI-5.2`, `UI-2.5`/`UI-4.1`, `UI-3.1`, `UI-5.5`, `UI-8.1`, `UI-9.3`) |
443> | REVIEW UI code or a design artifact | Each clause is a fail-condition; every finding cites `UI-<clause>` + `file:line` + severity. NEVER a tick-box sweep — one focused pass per section |
444> | SCAN / document a UI system | Record where the PROJECT deliberately deviates, so the overriding doc becomes the recorded authority |
445>
446> **Skip ONLY** when the change has no user-facing surface (backend-only, tooling, docs) — state that explicitly so the skip is auditable, not an omission.
447
448<!-- /SYNC:ui-ux-design-principles -->
449
450<!-- SYNC:ai-mistake-prevention -->
451
452> **AI Mistake Prevention** — Failure modes to avoid on every task:
453>
454> **Re-read files after context changes.** Context compaction, resume, or long-running work can make memory stale; verify current files before acting.
455> **Verify generated content against source evidence.** AI hallucinates APIs, names, claims, and document facts. Check the relevant source before documenting or referencing.
456> **Check downstream references before deleting or renaming.** Removing an artifact can stale docs, generated mirrors, configs, and callers; map references first.
457> **Trace the full impact chain after edits.** Changing a definition can miss derived outputs and consumers. Follow the affected chain before declaring done.
458> **Verify ALL affected outputs, not just the first.** One green check is not all green checks; validate every output surface the change can affect.
459> **Assume existing values are intentional — ask WHY before changing OR flagging one as a defect.** Before changing or reporting a constant, limit, flag, cutoff, wording, or pattern, read nearby context and history, the CALLER's ordering, and 2+ sibling call sites of the same convention. A doc stating WHAT without WHY is missing rationale, not proof of a missing guard.
460> **Surface ambiguity before acting — don't pick silently.** Multiple valid interpretations require an explicit question or stated assumption with risk.
461> **Assert the outcome your system owns, not the intermediate state your infrastructure owns.** When verifying async work, assert the final business state — never the delivery/retry bookkeeping held in shared infrastructure that any co-running process can write. Such a check passes when run alone and flakes the moment anything else shares that infrastructure.
462> **Keep shared guidance role-relevant.** Universal guidance must help every receiving skill or agent; code-specific obligations belong only in code-specific protocols.
463
464<!-- /SYNC:ai-mistake-prevention -->
465
466<!-- SYNC:critical-thinking-mindset -->
467
468> **Critical Thinking Mindset** — Apply critical thinking, sequential thinking. Every claim needs traced proof, confidence >80% to act.
469> **Anti-hallucination:** Never present guess as fact — cite sources for every claim, admit uncertainty freely, self-check output for errors, cross-reference independently, stay skeptical of own confidence — certainty without evidence root of all hallucination.
470
471<!-- /SYNC:critical-thinking-mindset -->
472
473<!-- SYNC:critical-thinking-mindset:reminder -->
474
475**MUST ATTENTION** apply critical + sequential thinking — every claim needs appropriate traced evidence (`file:line` for repo/code claims; source URL or artifact section for research, product, content, and docs claims); confidence >80% to act, <60% DO NOT recommend. Anti-hallucination: never present guess as fact, admit uncertainty freely, cross-reference independently, stay skeptical of own confidence.
476
477<!-- /SYNC:critical-thinking-mindset:reminder -->
478
479<!-- SYNC:ai-mistake-prevention:reminder -->
480
481**MUST ATTENTION** apply AI mistake prevention — verify generated content against evidence, trace downstream references before deleting or renaming, verify all affected outputs, re-read files after context loss, and surface ambiguity before acting.
482
483<!-- /SYNC:ai-mistake-prevention:reminder -->
484
485<!-- SYNC:ui-ux-design-principles:reminder -->
486
487- **MUST ATTENTION** apply the 40 UI/UX Design Principles (`UI-1.1`–`UI-9.4`) to any user-facing surface: one focal point, proximity grouping, empty/loading/error designed FIRST (§1) · ≤2 families, 16px web / 17px mobile body, never <14px, 45–75ch, fixed 6-step scale (§2) · 4.5:1 text / 3:1 edges measured, one accent, colour never alone, dark mode ≠ inversion (§3) · one 4/8px unit, `gap` over margins, tighter-inside-looser-between, content-driven breakpoints (§4) · <100ms response, all 5 states, undo over confirm, 150–250ms ease-out honouring reduced-motion, visible focus ring (§5) · where-am-I/what's-here/where-next, ≤5 top-level destinations, user's words, URL or back path (§6) · fewer fields, visible labels, validate on blur, matched keyboard, NEVER lose input (§7) · ≥44×44pt targets 8px apart, bottom-third primaries, gestures never the only route, safe areas (§8) · structure before data, optimistic update with visible rollback, reserved space, slow-connection states (§9). Project design-system docs OUTRANK these clauses — a genuine conflict goes to the user, NEVER resolved silently. Cite every finding as `UI-<clause>` + `file:line`. Skip ONLY for changes with no user-facing surface, stated explicitly.
488
489<!-- /SYNC:ui-ux-design-principles:reminder -->
490
491<!-- SYNC:project-protocol-overlay -->
492
493> **Project Protocol Overlay** — Before executing this skill, resolve any PROJECT overlay rules layered onto it: match this skill's name against the `Target` column of the project's skill-protocol index (`docs/project-reference/skill-protocols-reference.md` by default; a `referenceDocs` entry in `docs/project-config.json` overrides the path), taking the most specific matching tier ONLY — exact name > glob > `*`. **That precedence orders overlays against EACH OTHER, never against this skill.** Read ONLY the matched bodies, resolved as `<protocols-dir>/<Name>.md`; a row's Body link is display text, never a read path. A matched body that is missing or malformed is REPORTED and skipped — never reconstructed from the index Description. No index, or no match -> proceed with no overlay, silently. Full contract: `.claude/skills/project-skill-protocol/references/registry.md`.
494>
495> Overlays are **ADDITIVE ONLY**: they ADD rules on top of this skill's own protocol and NEVER replace, override, disable, or reinterpret a rule it already states — removing every overlay must return this skill to exactly its documented behavior. An overlay is a BRIEF, not an authority escalation: it can NEVER waive a workflow gate, git discipline, a review gate, or a user-confirmation gate. A genuine overlay-vs-skill conflict, or two equally-specific overlays that directly contradict -> surface both to the user; NEVER resolve silently.
496
497<!-- /SYNC:project-protocol-overlay -->
498
499<!-- SYNC:project-protocol-overlay:reminder -->
500
501**MUST ATTENTION** resolve project protocol overlays for this skill BEFORE executing — most specific matching tier only (exact > glob > `*`, which ranks overlays against each other, NEVER against this skill), read only matched bodies at `<protocols-dir>/<Name>.md`; a missing or malformed body is reported, never reconstructed. Overlays are ADDITIVE ONLY (they never replace this skill's own rules) and are a brief, NEVER an authority escalation; an equal-specificity contradiction goes to the user.
502
503<!-- /SYNC:project-protocol-overlay:reminder -->
504
505## Closing Reminders
506
507**IMPORTANT MUST ATTENTION Goal:** Provide searchable UI/UX design intelligence — 50 styles, 21 palettes, 50 font pairings, 20 charts, 8 stacks — synthesized into accessible, professional, contrast-correct interfaces.
508
509**Protocols in force (concise digest of the SYNC/shared blocks this skill carries):**
510
511- **AI Mistake Prevention:** verify generated content against evidence, trace downstream references, verify all affected outputs, re-read after context loss, surface ambiguity.
512- **Critical Thinking:** Traced `file:line` proof per claim, confidence >80% to act, never guess.
513
514**IMPORTANT MUST ATTENTION** break work into small todo tasks using task tracking BEFORE starting
515**IMPORTANT MUST ATTENTION** search codebase for 3+ similar patterns before creating new code
516**IMPORTANT MUST ATTENTION** cite `file:line` evidence for every claim (confidence >80% to act)
517**IMPORTANT MUST ATTENTION** add a final review todo task to verify work quality
518**IMPORTANT MUST ATTENTION** the 40 UI/UX Design Principles (`UI-1.1`–`UI-9.4`) CONSTRAIN the curation — a curated palette must still measure 4.5:1 text / 3:1 edges (`UI-3.1`) with one accent (`UI-3.2`) and colour never alone (`UI-3.3`); a curated font pairing must still be ≤2 families × 3 weights (`UI-2.1`) at 16px web / 17px mobile, never <14px (`UI-2.2`), delivered as a 6-step type scale (`UI-2.5`); spacing is ONE 4/8px unit multiplied (`UI-4.1`); anything built from it specifies empty/loading/error FIRST (`UI-1.5`) and all 5 interaction states (`UI-5.2`) — a search hit that fails a clause is a finding, NOT a deliverable
519
520**[TASK-PLANNING]** Before acting, analyze task scope and systematically break it into small todo tasks and sub-tasks using task tracking.
521
522<!-- CODEX:SYNC-PROMPT-PROTOCOLS:START -->
523
524## Hookless Prompt Protocol Mirror (Auto-Synced)
525
526Source: `.claude/.ck.json` + `.claude/skills/shared/sync-inline-versions.md` (`:full` blocks) + `.claude/scripts/lib/hookless-prompt-protocol.cjs`
527
528## [WORKFLOW-EXECUTION-PROTOCOL] [BLOCKING] Workflow Execution Protocol — MANDATORY IMPORTANT MUST CRITICAL. Do not skip for any reason.
529
530**Generic portability boundary:** Reusable skills and protocol text stay project-neutral; project-specific conventions are discovered from docs/project-config.json and docs/project-reference/. Apply shared AI-SDD from `shared/sdd-artifact-contract.md`. Read `docs/project-config.json` and `docs/project-reference/docs-index-reference.md`, then open the project reference docs named there. For spec, test-case, behavior-change, public-contract, or `docs/specs/` work, route through the local spec docs named by the docs index: `feature-spec-reference.md`, `spec-system-reference.md`, `spec-principles.md`, and `workflow-spec-test-code-cycle-reference.md` when specs/tests/code must stay synchronized. If either file or a required reference doc is missing or stale, auto-run `$project-init` (or the narrow lower-level route such as `$project-config`, `$docs-init`, `$scan-all`, or `$scan --target=<key>`) before ordinary project-specific work. Any supported AI tool may execute when this shared context and local docs are available.
531
5321. **DETECT:** If the prompt starts with an explicit slash skill/workflow command, execute it directly. Otherwise match the prompt against the workflow catalog and skill list.
5332. **ANALYZE:** Choose the best option: execute directly, invoke a skill, activate a standard workflow, or compose a custom step combination.
5343. **AUTO-SELECT:** Pick the best option yourself. Do not ask the user to choose between direct execution, skill, standard workflow, or custom workflow.
5354. **ACTIVATE:** For a selected workflow, call `$start-workflow <workflowId>`; for a selected skill, invoke that skill; for a custom workflow, sequence custom steps directly; for direct execution, proceed with the task.
5365. **CREATE TASKS:** task tracking for ALL workflow/skill/custom steps before execution when the selected path has multiple steps.
5376. **PARALLELIZE:** Before executing the task list, tag each task `PAR` (independent inputs + write set disjoint from every other `PAR` task) or `SEQ` (name the blocking dependency), group `PAR` tasks into waves, declare the wave plan, and spawn each wave's sub-agents in ONE message — all-return barrier per wave, fan-out one level deep unless a sub-agent's own definition authorizes further fan-out. Sequential-by-default is a defect when tasks are independent; do not parallelize shared write targets, output-consuming tasks, trivial single-file work, ordering a skill or workflow explicitly fixes, or user-approval gates.
5387. **EXECUTE:** Advance per the **Workflow Step Advancement & Parallel Phases** rule in your context instructions — model-driven; a sub-agent completion advances a step identically to an inline call; a parallel-phase group is an all-return barrier (advance only after ALL members return, never serialize it)
539
540## Shared AI-SDD Protocol Markers
541
542Source: `.claude/skills/shared/sync-inline-versions.md`
543
544## SYNC:ai-sdd-artifact-contract
545
546> **AI-SDD Artifact Contract** — Shared spec-driven development rules stay portable and source-owned.
547>
548> 1. Keep reusable AI-SDD principles in `.claude`; put repository-specific paths, commands, owners, products, and formats in project config/reference docs.
549> 2. Preserve cycle: `spec -> plan -> tasks -> implement -> verify -> update spec/docs`.
550> 3. Trace every requirement or invariant through decision, task, TC/test, source evidence, and docs/spec update.
551> 4. Treat code-to-spec extraction as reference-only until accepted by the canonical spec owner.
552> 5. Any supported AI tool may plan, implement, review, or verify with synced context; using multiple tools is optional.
553> 6. Update `.claude` source first, then sync generated mirrors; do not manually edit `.agents`, `.codex`, or `AGENTS.md`. — why: mirrors are generated artifacts; hand-edits are overwritten on the next sync
554> 7. If `docs/project-config.json`, root instruction files, or a required project-reference doc is missing or stale, auto-run `$project-init` or the narrow lower-level route before ordinary project-specific work.
555>
556> **Active reference:** `shared/sdd-artifact-contract.md` in the active skills root.
557
558---
559
560## SYNC:ai-sdd-artifact-contract:reminder
561
562- **MANDATORY** Apply `shared/sdd-artifact-contract.md`; keep reusable AI-SDD in `.claude` and local rules in project docs.
563- **MANDATORY** Code-to-spec extraction is reference-only until canonical acceptance; any supported AI tool may execute with synced context.
564- **MANDATORY** Update `.claude` source before syncing generated mirrors; do not manually edit `.agents`, `.codex`, or `AGENTS.md`.
565- **MANDATORY** Missing or stale project config, root instruction files, or required reference docs route project-specific work through `$project-init` or the narrow setup route automatically.
566 **[TASK-PLANNING] [MANDATORY]** BEFORE executing any workflow or skill step, create/update task tracking for all planned steps, then keep it synchronized as each step starts/completes.
567
568## [LESSON-LEARNED-REMINDER] [BLOCKING] Task Planning & Continuous Improvement — MANDATORY. Do not skip.
569
570Break work into small tasks (task tracking) before starting. Add final task: "Analyze AI mistakes & lessons learned".
571
572**Extract lessons — ROOT CAUSE ONLY, not symptom fixes:**
573
5741. Name the FAILURE MODE (reasoning/assumption failure), not symptom — "assumed API existed without reading source" not "used wrong enum value".
5752. Generality test: does this failure mode apply to ≥3 contexts/codebases? If not, abstract one level up.
5763. Write as a universal rule — strip project-specific names/paths/classes. Useful on any codebase.
5774. Consolidate: multiple mistakes sharing one failure mode → ONE lesson.
5785. **Recurrence gate:** "Would this recur in future session WITHOUT this reminder?" — No → skip `$learn`.
5796. **Auto-fix gate:** "Could `$code-review`/`$code-simplifier`/`$security-review`/`$lint` catch this?" — Yes → improve review skill instead.
5807. BOTH gates pass → ask user to run `$learn`.
581 **[CRITICAL-THINKING-MINDSET]** Apply critical thinking, sequential thinking. Every claim needs traced proof, confidence >80% to act.
582 **Anti-hallucination principle:** Never present guess as fact — cite sources for every claim, admit uncertainty freely, self-check output for errors, cross-reference independently, stay skeptical of own confidence — certainty without evidence root of all hallucination.
583 **AI Attention principle (Primacy-Recency):** Put the 3 most critical rules at both top and bottom of long prompts/protocols so instruction adherence survives long context windows.
584 **Goal-driven execution:** Define success criteria first, loop until verified, and stop only when observable checks pass.
585 **Tests verify intent:** Tests must protect business rules/invariants and fail when the protected intent breaks, not only mirror current behavior.
586
587## Common AI Mistake Prevention (System Lessons)
588
589- **Re-read files after context compaction.** Edit requires prior Read in same context; compaction wipes read state. Re-read before editing.
590- **Grep for old terms after bulk replacements.** AI over-trusts find/replace completeness. Grep full repo after bulk edits for missed refs in docs/configs/catalogs.
591- **Check downstream references before deleting.** Deletions cascade doc/code staleness. Map referencing files before removal.
592- **After memory loss, check existing state before creating new.** Compaction wipes prior-work memory. Query current state to resume — never blindly duplicate.
593- **Verify AI-generated content against actual code.** AI hallucinates APIs, class names, method signatures. Grep to confirm existence before documenting/referencing.
594- **Trace full dependency chain after edits.** Changing a definition misses downstream consumers. Trace the full chain.
595- **When renaming, grep ALL consumer file types.** Some file types silently ignore missing refs (no compile error). Search code, templates, configs, generated files.
596- **Trace ALL code paths when verifying correctness.** Code existing ≠ code executing. Trace early exits, error branches, conditional skips — not just happy path.
597- **Update docs that embed canonical data when source changes.** Docs inlining derived data (workflows, schemas, configs) go stale silently. Update all embedding docs alongside source.
598- **Verify sub-agent results after context recovery.** Background agents may finish while parent compacted — grep-verify output, don't trust assumed completion.
599- **Cross-check full target list against sub-agent assignments.** Parallel sub-agents by category miss boundary items. Reconcile union of assignments against target list before proceeding.
600- **Sub-agents inherit knowledge only from their agent .md definition — use custom agent types, not built-in Explore.** Tool adoption = permission + knowledge + enforcement (numbered workflow step).
601- **Persist sub-agent findings incrementally, not as a final batch.** Long sub-agents hit cutoffs before final write — findings lost. Instruct append-per-section to report file.
602- **When debugging, ask "whose responsibility?" before fixing.** Trace caller (wrong data) vs callee (wrong handling). Fix at responsible layer — never patch symptom site.
603- **Test failure → record a provisional verdict before trace/edit, then investigate.** Use the full five-way taxonomy: SOURCE-WRONG (production violates intent), TEST-WRONG (assertion/setup is stale), TEST-NOT-OPTIMAL (valid but fragile or low-signal test), ENVIRONMENT-BLOCKED (external state prevents a verdict), or AMBIGUOUS (intent/evidence cannot choose safely). Then trace root cause and triangulate against the governing spec (`docs/specs/**` if one exists) AND source. NEVER weaken an assertion, add a skip, relax a timeout, or change source merely to force green.
604- **Grep ALL removed names after extraction/refactoring.** Primary file "done" ≠ secondary files clean. Grep entire scope for every removed symbol before declaring complete.
605- **Assume existing values are intentional — ask WHY before changing OR flagging one as a defect.** Pattern-matching as "wrong" skips context. Before changing or reporting any constant/limit/flag/cutoff: read comments, git blame, the CALLER's ordering (the guarantee that makes the value correct usually lives in code running immediately BEFORE the cited line), and 2+ sibling call sites of the same convention. A doc stating WHAT without WHY is missing rationale, not proof of a missing guard — and in a validation pass, an accurate `file:line` citation proves the transcription, never the defect.
606- **Verify ALL affected outputs, not just the first.** One build green ≠ all green. Multi-stack changes (backend/frontend/tests/docs) require verifying EVERY output.
607- **Evaluate fit before copying a nearby pattern.** Closest example ≠ matching preconditions — verify the new context shares the same constraints, base classes, scope, lifetime.
608- **Holistic-first debugging — resist nearest-attention trap.** Don't dive into first plausible cause. List EVERY precondition (config, env vars, paths, DB, endpoints, creds, versions, DI, data). Verify each against evidence (grep/query — not reasoning). Ask "what would falsify this?" — if nothing, it's not a hypothesis. Most expensive failure: going deeper in "obvious" layer while bug sits in layer never questioned.
609- **Surgical changes — apply the diff test (context-aware).** Two modes: (1) Bug fix → every line traces to the bug; no restyling; orphan cleanup only for imports YOUR changes made unused. (2) Review/enhancement → implement improvements AND announce as "Enhancement beyond main request: [what]". Never silently scope-creep. Diff test: "Would this line exist if I wasn't asked to do X?" — if no, delete or announce.
610- **Surface ambiguity before coding — don't pick silently.** Multiple valid interpretations → present each with effort: "[Request] could mean (1) [N h], (2) [N h]. Which matters?" List scope/format/volume/constraints assumptions first. If simpler path exists, say so. Never silently pick.
611- **[MANDATORY FIRST ACTION] ALWAYS activate a suitable skill or workflow BEFORE responding.** Match task against workflow catalog + skill list; invoke via skill invocation or `$start-workflow <workflowId>`. NEVER answer or write code before checking. Skip = protocol violation.
612- **Why-Review adversarial mindset — apply when reviewing any plan, decision, or design.** Default SKEPTIC not VALIDATOR: steel-man a rejected alternative, invert each stated reason ("what does it sacrifice?"), stress-test top 2-3 assumptions, run pre-mortem ("ships, fails in 3 months — what breaks?"), surface 1-2 alternatives author missed. Section presence ≠ quality; quality = causal reasoning + concrete mitigations + evidence, not "it's better" or "monitor closely".
613- **Front-load report-write in sub-agent prompts for large reviews.** Many-file sub-agents hit budget before final write — findings lost. Design prompts so: (1) report-write is first explicit deliverable, (2) append per-file/section (not batched), (3) scope bounded so reads don't exhaust budget. Truncated mid-sentence with no report file → spawn narrower scope, don't retry same prompt.
614- **After context compaction, re-verify all prior phase outcomes before continuing.** Summaries describe intent, not environment state (git index, filesystem, processes). On resume, FIRST audit: git status, re-read modified files, verify filesystem. Every "completed" claim is an untested hypothesis until evidence confirms.
615- **OOM/memory: check row count before row size.** Triage: (1) Unbounded query — no DB filter for trigger? Push filter to DB; eliminates OOM. (2) Large rows? Projection reduces proportionally. Row reduction > projection in ROI.
616- **Assert the outcome your system OWNS, never the intermediate state your INFRASTRUCTURE owns.** When testing anything asynchronous (queue/broker delivery, retries, background jobs, caches, replication), assert the final business/entity state. NEVER assert the delivery bookkeeping — consume/send status, attempt counts, last-error, row existence or counts in a broker, scheduler, or outbox/inbox table. That bookkeeping lives in shared infrastructure that ANY co-running process (a peer worker, a second replica, a leftover local container) can write, usually under a deterministic shared key, so the assertion silently tests the developer's environment instead of the system: green when run alone, flaky the instant anything else shares that broker + database. Gate question for every assertion: "would this hold no matter WHICH process did the work?" — if no, assert the converged data state instead. Corollary: process-local fault injection and in-process telemetry cannot gate work any process may perform — use them as stress amplifiers (arm → bounded window → disarm → assert convergence), never as preconditions.
617- **Keep domain concepts out of generic/shared/infrastructure layers.** Reusable layer (shared library, framework, infra module) must reference NO consumer-specific domain concept — tenant/customer/product IDs, business entities, feature rules. Leak compiles + runs → passes review silently while coupling the "reusable" layer to one consumer. Keep shared type domain-free; push domain fields/logic down into the consumer via subclass/composition. — why: a layer coupled to one consumer's domain is no longer reusable.
618
619<!-- CODEX:SYNC-PROMPT-PROTOCOLS:END -->