1---
2name: design-spec
3description: '[Project Management] Use when you need to create UI/UX design specifications from requirements, PBIs, or user stories. Use --mode=wireframe to convert hand-drawn/digital wireframes or UI sketches into structured specs.'
4metadata:
5 source: https://skillmd.com/skills/duc01226/design-spec
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<!-- PROMPT-ENHANCE:STEP-TASK-ANCHOR:START -->
52
53> **[BLOCKING]** Execute skill steps in declared order. NEVER skip, reorder, or merge steps without explicit user approval.
54> **[BLOCKING]** Before each step or sub-skill call, update task tracking: set `in_progress` when step starts, set `completed` when step ends.
55> **[BLOCKING]** Every completed/skipped step MUST include brief evidence or explicit skip reason.
56> **[BLOCKING]** If Task tools are unavailable, create and maintain an equivalent step-by-step plan tracker with the same status transitions.
57
58<!-- PROMPT-ENHANCE:STEP-TASK-ANCHOR:END -->
59
60## Quick Summary
61
62**Goal:** Produce a structured, tech-agnostic UI/UX design specification from a requirement/PBI/story/wireframe so a developer can rebuild the UI on ANY stack — every component, state, design token, responsive rule, and accessibility need documented and linked back to the governing Feature Spec.
63
64**Summary (read this if nothing else — PURPOSE + every main step):**
65
66- **Purpose** — translate requirements into a developer-ready UI/UX spec (component inventory, states, design tokens, responsive behavior, accessibility), tech-agnostic: name components by UX role, never by framework/library class.
67- **Step 0–0b — research first:** inventory existing related screens/components + map connected flows, record in §1 so the spec matches the live UI system; if a governing Feature Spec exists, seed from its §6 interaction surface and reuse its view + observable-state vocabulary verbatim. — why: divergent vocabulary breaks the navigable spec↔design hub.
68- **Step 1–2 — read & route input, set complexity:** Figma URL → `$figma-design`; image/screenshot → visual analysis tooling; wireframe/sketch → `--mode=wireframe`; PBI/text → extract requirements. Pick Quick Spec (§1–4) vs Full Spec (§1–7, +Flow Diagram for multi-page).
69- **Step 3–6 — author the body:** build component inventory (new vs existing), define all 7 states + interactions, extract design tokens (reuse design-system), document responsive breakpoints (mobile 320–767 / tablet 768–1023 / desktop 1024+).
70- **Step 7–8 — save & link back:** write the artifact to `team-artifacts/design-specs/`, then set the governing Feature Spec's `design_spec:`/`mockup:` frontmatter to the saved path — frontmatter only, never the §1–8 body. — why: `artifact-review --type=design` fails if the path is not recorded.
71
72**Workflow:**
73
741. **Read Source** — Extract UI requirements from PBI, story, or Figma URL
752. **Determine Complexity** — Quick Spec (sections 1-4) vs Full Spec (all 7 sections)
763. **Build Component Inventory** — List new vs existing components
774. **Define States & Tokens** — Interactions, design tokens, responsive breakpoints
785. **Save Artifact** — Output to `team-artifacts/design-specs/`
79
80**Key Rules:**
81
82- If Figma URL provided → auto-routes to `$figma-design` for context extraction
83- If wireframe image provided (hand-drawn/digital/tool-export) → handled internally via `--mode=wireframe` (see "Mode: wireframe" below)
84- If screenshot provided → uses `visual analysis tooling` for design extraction
85- Reference existing design system tokens from `docs/project-reference/design-system/`
86- Component patterns: `docs/project-reference/frontend-patterns-reference.md`
87- Include accessibility requirements (keyboard nav, ARIA labels, contrast)
88- **[BLOCKING] Tech-agnostic output:** spec prose/headings follow `docs/project-reference/spec-principles.md` §3 — describe components by UX role, not framework/library names; source paths and class names appear ONLY in evidence fields (`**Evidence**`, `[Source:]`), frontmatter, and Mermaid.
89
90**Be skeptical. Apply critical thinking, sequential thinking. Every claim needs traced proof, confidence percentages (Idea should be more than 80%).**
91
92# Design Specification
93
94Create structured UI/UX design specification documents from requirements or PBIs for developer handoff.
95
96## When to Use
97
98- A PBI or user story needs a design spec before implementation
99- Translating requirements into concrete UI layout, states, and tokens
100- Documenting component inventory and interaction patterns
101- Creating responsive breakpoint specifications
102
103## When NOT to Use
104
105- This skill auto-routes Figma URLs to `$figma-design`; wireframes are handled internally via `--mode=wireframe` — no need to call a separate skill
106- Building the actual UI -- use `design --lane=marketing` (marketing/creative) or `design --lane=product` (product UIs)
107- Reviewing existing UI code -- use `web-design-guidelines`
108
109## Prerequisites
110
111Read before executing:
112
113- The source PBI, user story, or requirements document
114- `docs/project-reference/design-system/` -- project design tokens (if applicable)
115- Existing design specs in `team-artifacts/design-specs/` for format consistency
116
117### Frontend/UI Context
118
119> When this task involves frontend or UI changes,
120
121- Frontend patterns: `docs/project-reference/frontend-patterns-reference.md`
122- Styling/BEM guide: `docs/project-reference/scss-styling-guide.md`
123- Design system tokens: `docs/project-reference/design-system/README.md`
124
125## Workflow
126
127> **[BLOCKING] Step 0 — Inventory existing UI + map connected flows** (per the `SYNC:existing-ui-research` protocol carried by this skill). Before authoring the spec, inventory the existing related screens / components / pages already serving this feature or domain, and map every connected feature flow that links to / embeds / navigates to-or-from the new screen. Record the matched screens + flows in §1 Overview so the spec faithfully matches the current UI system. Skip only for backend-only work (state it explicitly).
128
129> **[BLOCKING] Step 0b — Seed from the governing Feature Spec's §6 interaction surface (when one exists).** Detect whether a canonical Feature Spec already governs this feature (look under `docs/specs/**` for a spec covering the same capability). If one exists, READ its **§6 Process Flows & Interaction Surface** — the **View Inventory** (§6.2), **Navigation Map** (§6.3), **Key UI States** (§6.4), and **Per-Story Interaction Flow** (§6.5) — and use it as this design-spec's starting frame:
130>
131> - **Reuse the spec's vocabulary verbatim** — carry over the SAME UX-role view names from §6.2 and the SAME observable-state names from §6.4 (the Default / Loading / Disabled / Error / Empty / Success vocabulary in this skill's "Observable State Definitions" table is the shared language). NEVER rename or re-partition what the spec already named — the two artifacts MUST speak the same language or the navigable hub breaks.
132> - **Deepen, do not diverge or contradict.** The spec stays tech-agnostic; this design-spec is the companion that adds visual fidelity (layout, tokens, pixel detail) ON TOP of the spec's intent. Map each §6.5 step and each §6.4 state into concrete visual treatment, preserving the `US-`/`OP-`/`BR-` logical-ID cross-refs the spec already owns.
133> - This coupling is the `SYNC:ui-intent-layer` contract carried below — see that block for the full rule; do not restate it here.
134>
135> **Skip ONLY** when no governing Feature Spec exists (author the interaction frame from the source PBI/story instead) — state which case applies.
136
1371. **Read source input & route by type**
138
139 | Input Detected | Detection | Action |
140 | ------------------------ | ---------------------------------------------- | ------------------------------------------------------------------------- |
141 | Figma URL | `figma.com/design` or `figma.com/file` in text | Activate `$figma-design` to extract context, then continue |
142 | Image/screenshot | Image file attached to prompt | Use `visual analysis tooling` to extract design guidelines, then continue |
143 | Hand-drawn wireframe | Image + "wireframe"/"sketch" keyword | Run `--mode=wireframe` (internal — see "Mode: wireframe" section) |
144 | PBI/story text | Acceptance criteria present | Extract UI requirements from text, continue |
145 | Verbal/text requirements | No image, no URL, no PBI | Clarify with user, then continue |
146
147For ANY visual input: extract design context FIRST, then proceed to spec generation.
148
1492. **Determine spec complexity**
150
151 ```
152 IF single form or simple component → Quick Spec (sections 1-4 only)
153 IF full page or multi-component view → Full Spec (all 7 sections)
154 IF multi-page flow → Full Spec + Flow Diagram
155 ```
156
1573. **Build component inventory**
158 - List all UI components needed
159 - Identify reusable vs feature-specific components
160 - Note existing components from shared component library or design system
161
1624. **Define states and interactions**
163 - Default, hover, active, disabled, error, loading, empty states
164 - User interactions (click, drag, keyboard shortcuts)
165 - Transitions and animations
166
1675. **Extract design tokens**
168 - Colors, typography, spacing, shadows, border-radius
169 - Reference existing design system tokens where possible
170
1716. **Document responsive behavior**
172 - Mobile (320-767px), Tablet (768-1023px), Desktop (1024px+)
173 - What changes at each breakpoint (layout, visibility, sizing)
174 - **Small-screen minimum bar (spec it explicitly):** the layout MUST stay usable on mobile. Preferred = reflow (rows `flex-wrap` / `row → column`, grids collapse to one column). Where a component genuinely can't reflow (data tables, canvases, wide grids), specify a `min-width`/`min-height` + `overflow: auto` scroll as the accepted fallback — scrolling is OK. Hard requirement = nothing broken (no clipped, cut-off, or unreachable content/controls). If a component needs a large redesign to work on mobile, flag it for the user rather than assuming a rewrite.
175
1767. **Save artifact** — pick the filename variant by artifact type:
177 - Design spec: `team-artifacts/design-specs/{YYMMDD}-designspec-{feature-slug}.md`
178 - Accessibility audit: `team-artifacts/design-specs/{YYMMDD}-ux-audit-{feature-slug}.md`
179 - Single-component doc: `team-artifacts/design-specs/{YYMMDD}-ux-component-{component-name}.md`
180
1818. **Link back to the governing Feature Spec (when one exists).** After saving the artifact, keep the spec the navigable hub: open the governing Feature Spec under `docs/specs/**` and set its frontmatter `design_spec:` key to this design-spec's saved path (add the key if absent, update it if stale). If a mockup was also produced (e.g. via `$pbi-mockup`), set the `mockup:` key the same way. Edit **frontmatter only** — never touch the §1–§8 spec body. This satisfies the `artifact-review --type=design` link-back gate, which fails when a design-spec exists but its path is not recorded in the spec's `design_spec:` frontmatter. Skip ONLY when no governing Feature Spec exists (the design-spec is standalone) — state that.
182
183### Role Context & Artifact Path (canonical)
184
185> Applies to Writes under `team-artifacts/design-specs/`.
186
187- **Active Role:** ui-ux-designer · **Skill:** design-spec
188- **Path:** `team-artifacts/design-specs/` · **Type:** designspec · **Role token:** ux
189- **Template:** `.claude/docs/team-artifacts/templates/design-spec-template.md`
190- **Naming:** `{YYMMDD}-ux-{type}-{slug}.md` (general artifact-path pattern: `{YYMMDD}-{role}-{type}-{slug}.md`)
191- **Context:** DESIGN SPEC — include component states, design tokens, accessibility requirements.
192- **Quality checklist:** `- [ ]` All states documented · `- [ ]` Design tokens specified · `- [ ]` Accessibility notes included · `- [ ]` Responsive breakpoints defined
193
194## Mode: wireframe (image → spec)
195
196> **Invoke with `--mode=wireframe`** (or whenever a hand-drawn wireframe, digital wireframe, or UI sketch is the input). This mode is an INPUT adapter: it analyzes the image, then flows into the normal spec sections (Output Format) and the M1-M5/M7 compliance gate. `design-spec` is the canonical owner of wireframe→spec conversion.
197
198### Input Routing (wireframe)
199
200| Input | Detection | Action |
201| ----------------------- | --------------------------------------- | -------------------------------------------- |
202| Hand-drawn sketch photo | Image with rough/organic lines | Analyze with wireframe prompts (this mode) |
203| Digital wireframe | Image with clean lines/shapes | Analyze with wireframe prompts (this mode) |
204| Wireframe tool export | Image from Excalidraw/Balsamiq/MockFlow | Analyze with wireframe prompts (this mode) |
205| Figma URL | `figma.com` in text | Route to `$figma-design` instead |
206| App screenshot | Polished UI with real data | Route to `$design --mode=screenshot` instead |
207
208### Wireframe Analysis
209
210Use `visual analysis tooling` with these prompts:
211
212**Prompt 1: Layout Extraction** — "Analyze this wireframe image. Identify: (1) page layout regions (header, sidebar, main, footer), (2) all UI elements with approximate position and type (button, input, table, card, dropdown, modal, tabs), (3) content hierarchy (what is primary vs secondary), (4) interactive elements, (5) any text labels or annotations, (6) navigation patterns."
213
214**Prompt 2: Component Identification** — "From the wireframe, list every distinct UI component. For each: name it descriptively, classify its complexity (primitive=single element, composite=grouped elements, section=page region), note its purpose."
215
216### Wireframe Output Generation
217
218After image analysis, generate (per the `SYNC:ui-wireframe-protocol` block below):
219
2201. **ASCII Wireframe** — Recreate layout using box-drawing characters
2212. **Component Inventory** — List with tier classification (Common/Domain-Shared/Page)
2223. **States Table** — Default, Loading, Empty, Error per view
2234. **Component Decomposition Tree** — If detail level warrants (refine/story)
2245. **Responsive Suggestions** — Based on layout complexity
225
226Apply the **M1-M5/M7 Compliance for UI Specs** gate (below) to all wireframe-derived prose: business-level component names, no code-prop refs, map to feature logic by logical ID, observable state transitions, rebuild-from-spec, business-visible subject matter.
227
228Wireframe-derived specs carry the same **Design-Principles Obligations** (below): the States Table (item 3) is authored empty/loading/error FIRST (`UI-1.5`) and covers all 5 interaction states per interactive element (`UI-5.2`); the Responsive Suggestions (item 5) break where the CONTENT breaks, not at device names (`UI-4.4`); and where the sketch is silent on type scale, spacing unit, or contrast (`UI-2.5`, `UI-4.1`, `UI-3.1`), record them in §4 as `[UNVERIFIED — needs design-system mapping]` rather than inventing one-off values measured off the drawing.
229
230### Mapped Business Operations
231
232Emit this table linking each interactive component to the feature operations/rules it drives (logical ID is the primary spine; mark `[UNVERIFIED — needs feature-spec mapping]` when the wireframe alone cannot determine it):
233
234| Interactive Component | Interaction (observable) | Feature Operation / Rule (logical ID) | Notes |
235| --------------------- | ------------------------ | ------------------------------------- | -------------------------------- |
236| Primary Button | Click → submit form | OP-XX | Triggers create/update operation |
237| Filter Dropdown | Select → reload list | OP-XX | Drives query/search operation |
238| Row Action Menu | Click → confirm dialog | BR-XX | Guarded by authorization rule |
239
240### Wireframe Output Formats
241
242- **Format A: PBI Section (default)** — output a standalone `## UI Layout` section compatible with PBI/story templates (consumed by `$pbi-mockup`).
243- **Format B: Standalone Spec** — output to `team-artifacts/design-specs/{YYMMDD}-wireframe-spec-{slug}.md`.
244
245### Confidence & Review (wireframe)
246
247- **Always display confidence level** for wireframe interpretation (analysis is 70-80% accurate).
248- **Always recommend human review** before proceeding to implementation.
249- If confidence <70%: ask clarifying questions about ambiguous elements by asking the user directly.
250
251## Output Format
252
253```markdown
254# Design Spec: {Feature Name}
255
256**Source:** {PBI/story reference}
257**Date:** {YYMMDD}
258**Status:** Draft | Review | Approved
259
260## 1. Overview
261
262{1-2 sentence summary of what this UI does}
263
264## 2. Component Inventory
265
266| Component | Type | Source | Notes |
267| --------- | -------- | ---------------- | --------------------------- |
268| UserCard | New | Feature-specific | Displays user avatar + name |
269| DataTable | Existing | shared library | Reuse with custom columns |
270
271## 3. Layout
272
273{Description or ASCII wireframe of layout structure}
274
275- Desktop: {layout description}
276- Tablet: {layout changes}
277- Mobile: {layout changes}
278
279## 4. Design Tokens
280
281| Token | Value | Usage |
282| ---------- | -------------- | --------------------- |
283| $primary | #1976D2 | Action buttons, links |
284| $text-body | 14px/1.5 Inter | Body text |
285| $gap-md | 16px | Section spacing |
286
287## 5. States & Interactions
288
289| Element | Default | Hover | Active | Disabled | Error |
290| -------- | ---------- | ---------- | ---------- | ---------------- | ----- |
291| Save btn | Blue/white | Darken 10% | Scale 0.98 | Gray/50% opacity | -- |
292
293## 6. Accessibility
294
295- Keyboard navigation order
296- ARIA labels for interactive elements
297- Color contrast compliance notes
298
299## 7. Open Questions
300
301- {Any unresolved design decisions}
302```
303
304## Design-Principles Obligations (per spec section)
305
306The spec is where the 40 UI/UX Design Principles (`UI-1.1`–`UI-9.4`, the `SYNC:ui-ux-design-principles` block below) become INHERITABLE — a downstream implementer builds what the spec SAYS, so an unstated clause is an unbuilt clause. Every spec this skill emits MUST carry these obligations in the named Output Format sections. Project design-system docs OUTRANK the clauses; a genuine conflict is surfaced to the user with both sides, NEVER resolved silently.
307
308| Spec section | Clause obligation the section MUST carry |
309| ----------------------------- | ------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------ |
310| **§1 Overview** | Name the ONE focal point of the screen (`UI-1.1`), and name the surface(s) in scope (web / mobile / both) so the mobile clauses (`UI-8.1`–`UI-8.4`) are either binding or explicitly skipped. |
311| **§3 Layout** | Group by proximity and shared alignment edges, never by nested borders (`UI-1.3`, `UI-1.4`); justify each breakpoint by where the CONTENT stops working, not by a device name (`UI-4.4`). |
312| **§4 Design Tokens** | **DECLARE the type scale as 6 named steps with no one-off sizes** (`UI-2.5`): body 16px web / 17px mobile and NEVER below 14px (`UI-2.2`), measure 45–75 characters (`UI-2.3`), leading 1.5 body / 1.1–1.2 display (`UI-2.4`), max 2 families × 3 weights (`UI-2.1`). **DECLARE ONE spacing unit** — 4px or 8px base with every gap a multiple (`UI-4.1`), space owned by the container via a gap property (`UI-4.2`), inner padding tighter than the gap to the next group (`UI-4.3`). A token table of ad-hoc values instead of a declared scale FAILS this gate. |
313| **§4 Design Tokens (colour)** | State the contrast TARGET AND THE MEASURED VALUE for every token pair — 4.5:1 text, 3:1 UI edges (`UI-3.1`) — measured, never eyeballed. One accent with one job (`UI-3.2`); dark mode specified as lifted surfaces + softened white text, NOT an inversion (`UI-3.4`). |
314| **§5 States & Interactions** | **Author the empty, loading and error state FIRST — before the populated state (`UI-1.5`).** Enumerate ALL 5 interaction states per interactive element — default, hover, focus, active, disabled — plus loading where it applies (`UI-5.2`, consistent with the 7-state Component States Checklist below). Response under 100ms even when the result takes longer (`UI-5.1`); motion 150–250ms ease-out honouring reduced-motion (`UI-5.4`); undo preferred over confirmation, confirm ONLY the irreversible (`UI-5.3`). Forms: labels always visible, placeholders are hints NEVER labels (`UI-7.2`); validate on blur with the fix-it message beside the field (`UI-7.3`); NEVER lose entered data across errors, navigation or refresh (`UI-7.5`). |
315| **§6 Accessibility** | The focus ring stays VISIBLE — restyled if it clashes, NEVER removed (`UI-5.5`); colour is never the sole carrier of meaning (`UI-3.3`); on any mobile/touch surface hit targets ≥44×44pt and 8px apart (`UI-8.1`), primary actions in the bottom third (`UI-8.2`), gestures never the only route (`UI-8.3`), safe areas and the on-screen keyboard respected (`UI-8.4`). |
316| **§7 Open Questions** | Every clause deliberately deviated from, with the project doc or user decision that authorises it — an undocumented deviation is a spec defect, not a style choice. |
317
318Keep every clause obligation written in observable UX language so it survives the M1-M5/M7 gate below — state the visible outcome, never a CSS property or framework prop.
319
320**Skip ONLY** when the feature has no user-facing surface — state that explicitly so the skip is auditable, not an omission.
321
322## M1-M5/M7 Compliance for UI Specs
323
324See `.claude/skills/shared/sdd-artifact-contract.md` → "AI-SDD Mandates (M1-M7)" for BLOCKING criteria. A UI spec MUST satisfy these before handoff:
325
326- **M1 — Business-level component names.** Name every component by its UX role — Primary Button, Secondary Button, Modal Dialog, Data Table, Dropdown, Toast — NEVER by a framework component class name or library import. FAIL on tech-term prose.
327- **M2 — No code-prop refs in prose.** Describe behavior and appearance in plain UX language. NEVER reference component-state props, CSS class names, framework directives, or selectors in prose. Those belong only in `**Evidence**`/`[Source:]` carriers, frontmatter, and Mermaid.
328- **M3 — Cross-reference by logical ID.** For every behavior driven by feature logic, cite the driving operation or rule by its logical ID (`OP-`/`BR-`/`FR-`) — link UI behavior back to the feature spec, not to handler code. Keep any `[Source: namespace/service/id]` abstract anchor strictly in the Evidence carrier — never physical code coordinates or repository-root paths.
329- **M4 — Testable, unambiguous behavior.** Every state and interaction MUST have exactly one valid interpretation and an observable completion marker. Replace vague phrases ("handle appropriately", "show feedback") with the concrete observable result.
330- **M5 — Rebuild-from-spec.** A reader with zero codebase knowledge MUST be able to rebuild this UI on ANY framework from the spec alone. If a marker is only resolvable by reading source, it fails M5 — restate it as a visual/textual observable.
331- **M7 — Business-visibility.** A UI spec is a business-tree artifact, so apply the demo test to each case's BODY: _"what would a stakeholder SEE change?"_ — no answer → FAIL as TECHNICAL-ONLY. Every `Given` = a state a user could arrange; every `When` = an action a user could take; every `Then` = an outcome a user could see. FAIL a `When` that is an invocation (a handler runs, a consumer receives, a job fires, data syncs) or a `Then` asserting schema/type/nullability/call-count. Judge the BODY, never the title or ID. ⚠️ **A UI state a user can SEE is business and PASSES M7** — a spinner, an empty placeholder, an error border, a disabled control are all demoable outcomes, not technical cases. Only invocation-shaped or schema-asserting cases fail.
332
333> **M1 vs M7 — they are not the same gate.** M1 governs **vocabulary**; M7 governs **subject matter**. A technical case in impeccably tech-free prose satisfies M1 while violating M7 — _"the view correctly reflects the synchronized record"_ names no framework, passes M1, and is still a technical case wearing a business costume. That gap is the most common way business specs rot. Ask what a user could SEE, not which words were used.
334
335### Observable State Definitions
336
337Define every state by what a user can SEE (color, icon, position, text), the business meaning, and the operation/rule that triggers it — NEVER by CSS class or component-state prop:
338
339| State | Visual Markers (observable) | Business Meaning | Triggering Operation / Rule (logical ID) |
340| -------- | ------------------------------------------------------ | ------------------------------------------ | ---------------------------------------- |
341| Default | Primary fill color, enabled label, no spinner | Action available to the actor | OP-XX (entry state) |
342| Loading | Spinner icon replaces label, control non-interactive | Operation in progress, awaiting result | OP-XX (request submitted) |
343| Disabled | Muted/gray fill, label dimmed, no pointer affordance | Precondition not met / actor not permitted | BR-XX (authorization or guard rule) |
344| Error | Error-color border, inline message text, alert icon | Operation rejected or validation failed | BR-XX (validation rule) |
345| Empty | Placeholder illustration + guidance text, no data rows | No records exist for the current view | OP-XX (query returned zero results) |
346| Success | Confirmation toast/checkmark, updated visible data | Operation completed and persisted | OP-XX (operation succeeded) |
347
348## Component States Checklist
349
350Every interactive component MUST document all 7 states by their **observable appearance and business meaning** — never by CSS class or framework prop (see M2):
351
352- **Default** — resting appearance; action available to the actor
353- **Hover** — pointer-over affordance change (cursor / elevation / color shift)
354- **Active** — pressed/engaged feedback during interaction
355- **Focus** — keyboard-focus indicator (visible ring/outline) for a11y traversal
356- **Disabled** — muted/non-interactive; precondition or permission not met
357- **Error** — validation/operation failure with inline message + alert affordance
358- **Loading** — in-progress indicator (spinner / skeleton); control non-interactive
359
360## Accessibility Audit (WCAG 2.1 AA)
361
362For an accessibility-audit deliverable, produce this checklist report and save it as `{YYMMDD}-ux-audit-{feature-slug}.md`:
363
364```markdown
365## Accessibility Audit: {Feature}
366
367**Date:** {Date}
368**Auditor:** {Name}
369**Standard:** WCAG 2.1 AA
370
371### Criteria Checklist
372
373#### Perceivable
374
375- [ ] 1.1.1 Non-text Content: Alt text for images
376- [ ] 1.3.1 Info and Relationships: Semantic HTML
377- [ ] 1.3.2 Meaningful Sequence: Logical reading order
378- [ ] 1.4.1 Use of Color: Not sole means of conveying info
379- [ ] 1.4.3 Contrast (Minimum): 4.5:1 text, 3:1 large text
380- [ ] 1.4.4 Resize Text: Readable at 200% zoom
381- [ ] 1.4.11 Non-text Contrast: 3:1 for UI components
382
383#### Operable
384
385- [ ] 2.1.1 Keyboard: All functions keyboard accessible
386- [ ] 2.1.2 No Keyboard Trap: Can navigate away
387- [ ] 2.4.1 Bypass Blocks: Skip navigation available
388- [ ] 2.4.3 Focus Order: Logical tab sequence
389- [ ] 2.4.4 Link Purpose: Clear from link text
390- [ ] 2.4.6 Headings and Labels: Descriptive
391- [ ] 2.4.7 Focus Visible: Clear focus indicator
392
393#### Understandable
394
395- [ ] 3.1.1 Language of Page: lang attribute set
396- [ ] 3.2.1 On Focus: No unexpected context change
397- [ ] 3.2.2 On Input: No unexpected context change
398- [ ] 3.3.1 Error Identification: Clear error messages
399- [ ] 3.3.2 Labels or Instructions: Form labels present
400
401#### Robust
402
403- [ ] 4.1.1 Parsing: Valid HTML
404- [ ] 4.1.2 Name, Role, Value: ARIA where needed
405
406### Issues Found
407
408| # | Criterion | Issue | Severity | Recommendation |
409| --- | --------- | ----- | -------- | -------------- |
410| 1 | | | P1/P2/P3 | |
411
412### Audit Status: PASS / FAIL / CONDITIONAL
413
414**Remediation Priority:**
415{List items by severity}
416```
417
418## Examples
419
420### Example 1: Simple form spec
421
422**Input:** "Design spec for customer onboarding form"
423
424**Output:** Quick Spec with sections 1-4 covering form fields (name, email, company name, plan-tier dropdown), validation rules, submit/cancel actions, and mobile stacking behavior.
425
426### Example 2: Complex dashboard spec
427
428**Input:** "Design spec for order pipeline dashboard with drag-and-drop columns"
429
430**Output:** Full Spec covering Kanban board layout, order cards (component inventory), drag-and-drop interactions, column states (empty, populated, over-limit), filter bar, responsive collapse to list view on mobile, and accessibility for keyboard drag operations.
431
432## Related Skills
433
434| Skill | When to use instead |
435| ------------------------- | ----------------------------------- |
436| `figma-design` | Extract specs from Figma designs |
437| `design --lane=marketing` | Build marketing/creative UI |
438| `design --lane=product` | Build product UI (dashboards, apps) |
439| `web-design-guidelines` | Review existing UI for compliance |
440
441---
442
443## Workflow Recommendation
444
445> **MANDATORY IMPORTANT MUST ATTENTION — NO EXCEPTIONS:** If you are NOT already in a workflow, you MUST ATTENTION use ask the user directly to ask the user. Do NOT judge task complexity or decide this is "simple enough to skip" — the user decides whether to use a workflow, not you:
446>
447> 1. **Run the design sequence** (Recommended) — `$design-spec` → `$design --lane=product` (product UIs) or `$design --lane=marketing` (marketing/creative) → `$workflow-review-changes`
448> 2. **Execute `$design-spec` directly** — run this skill standalone
449
450---
451
452> **[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.
453
454<!-- SYNC:existing-ui-research -->
455
456> **[BLOCKING] Understand the existing UI before you design or spec a new/updated screen.** Before producing any wireframe, mockup, screen design, or UI spec:
457>
458> 1. **Inventory existing related UI** — search the project for screens, pages, and components already serving this feature or its domain (consult design-system docs + the real component inventory).
459> 2. **Map connected flows** — identify every feature that links to, embeds, includes, or navigates to/from the new screen; trace its entry and exit flows so the new screen fits them.
460> 3. **Reuse before invent** — prefer existing components, patterns, and layout conventions; justify any new component against what already exists.
461> 4. **Record findings** — note the matched existing screens/components + connected flows in the artifact so downstream design faithfully matches the current UI system.
462>
463> **Skip ONLY** when the feature is backend-only (no UI) — state that explicitly.
464
465<!-- /SYNC:existing-ui-research -->
466
467<!-- SYNC:ui-intent-layer -->
468
469> **[BLOCKING] Capture a tech-agnostic UI/UX intent layer in every UI-bearing spec — a reader must be able to visualize how the feature works without naming any technology.** When the feature has a user interface, the spec MUST ATTENTION carry an interaction-surface section so the application — not just its API — can be rebuilt on any stack:
470>
471> 1. **View Inventory** — list each view/screen by its UX ROLE and purpose (e.g. "list of items", "item editor", "confirmation step") and what information it presents. Describe by role, never by an implementation name.
472> 2. **Navigation Map** — how a user moves between views: entry points, transitions, and exits. Trace how this surface connects to neighboring features already in the system.
473> 3. **Key observable UI States** — the distinct states a user can observe per view (empty, loading, populated, error, success, permission-denied, etc.) — described as what the user perceives, not how it is rendered.
474> 4. **Per-story interaction flow** — for each user story, the step-by-step click/action path from intent to outcome, cross-referenced to the logical IDs the spec already owns (`US-`/`OP-`/`BR-`).
475> 5. **Couple to the companion design artifact** — keep deep visual fidelity (layout, tokens, pixel detail) OUT of the spec; it lives in the linked `design-spec`/mockup. Record that companion's path in the spec frontmatter so the spec stays the navigable hub.
476>
477> **M1-clean (NON-NEGOTIABLE):** the prose names ZERO frameworks, routes/URLs, CSS, or component-class names — only roles, information, states, and flows. Technology detail belongs in the companion design artifact, never here.
478>
479> **Skip ONLY** when the feature is backend-only (no UI) — state that reason explicitly in the section.
480
481<!-- /SYNC:ui-intent-layer -->
482
483<!-- SYNC:ai-mistake-prevention -->
484
485> **AI Mistake Prevention** — Failure modes to avoid on every task:
486>
487> **Re-read files after context changes.** Context compaction, resume, or long-running work can make memory stale; verify current files before acting.
488> **Verify generated content against source evidence.** AI hallucinates APIs, names, claims, and document facts. Check the relevant source before documenting or referencing.
489> **Check downstream references before deleting or renaming.** Removing an artifact can stale docs, generated mirrors, configs, and callers; map references first.
490> **Trace the full impact chain after edits.** Changing a definition can miss derived outputs and consumers. Follow the affected chain before declaring done.
491> **Verify ALL affected outputs, not just the first.** One green check is not all green checks; validate every output surface the change can affect.
492> **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.
493> **Surface ambiguity before acting — don't pick silently.** Multiple valid interpretations require an explicit question or stated assumption with risk.
494> **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.
495> **Keep shared guidance role-relevant.** Universal guidance must help every receiving skill or agent; code-specific obligations belong only in code-specific protocols.
496
497<!-- /SYNC:ai-mistake-prevention -->
498
499<!-- SYNC:ui-system-context -->
500
501> **UI System Context** — For ANY task touching `.ts`, `.html`, `.scss`, or `.css` files:
502>
503> **MUST ATTENTION READ before implementing:**
504>
505> 1. `docs/project-reference/frontend-patterns-reference.md` — component base classes, stores, forms
506> 2. `docs/project-reference/scss-styling-guide.md` — BEM methodology, SCSS variables, mixins, responsive
507> 3. `docs/project-reference/design-system/README.md` — design tokens, component inventory, icons
508>
509> Reference `docs/project-config.json` for project-specific paths.
510
511<!-- /SYNC:ui-system-context -->
512
513<!-- SYNC:ui-ux-design-principles -->
514
515> **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`.
516>
517> **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.
518>
519> **1.0 Visual Hierarchy & Layout**
520>
521> - `UI-1.1` One focal point per screen. Two elements competing for first read → demote one.
522> - `UI-1.2` Signal order: size → weight → colour → position. Use the cheapest signal that works before adding another.
523> - `UI-1.3` Group by proximity, NEVER by border. Whitespace separates cleanly; boxes inside boxes do not.
524> - `UI-1.4` Align to a shared edge. Every unexplained indent reads as an accident.
525> - `UI-1.5` Design the empty, loading and error state FIRST. The full state is the easy one.
526>
527> **2.0 Typography**
528>
529> - `UI-2.1` Max 2 families, 3 weights each. More variety reads as inconsistency, not range.
530> - `UI-2.2` Body text 16px web, 17px mobile. NEVER below 14px for anything a user must read.
531> - `UI-2.3` Line length 45–75 characters. Constrain the measure, not the container.
532> - `UI-2.4` Leading scales inversely with size: 1.5 body, 1.1–1.2 display.
533> - `UI-2.5` Fixed type scale — 6 named steps shared with engineering. NEVER one-off sizes.
534>
535> **3.0 Colour & Contrast**
536>
537> - `UI-3.1` Contrast 4.5:1 text, 3:1 UI edges. Measure it — NEVER judge by eye on a bright screen.
538> - `UI-3.2` One accent, one job. An accent that is everywhere points at nothing.
539> - `UI-3.3` Colour NEVER carries meaning alone — pair it with an icon, label or position.
540> - `UI-3.4` Dark mode is NOT inverted light mode. Lift surfaces to signal elevation; soften pure-white text.
541>
542> **4.0 Spacing & Grid**
543>
544> - `UI-4.1` One spacing unit, multiplied — 4px or 8px base; every gap a multiple of it.
545> - `UI-4.2` Space belongs to the container, not the child. Use `gap`; reserve margins for exceptions.
546> - `UI-4.3` Tighter inside, looser between — inner padding always smaller than the gap to the next group.
547> - `UI-4.4` Breakpoints follow content, not devices. Break where the layout stops working.
548>
549> **5.0 Interaction & Feedback**
550>
551> - `UI-5.1` Every action gets a response under 100ms, even when the result takes longer.
552> - `UI-5.2` Specify all 5 states — default, hover, focus, active, disabled — plus loading where it applies.
553> - `UI-5.3` Prefer undo over confirmation. Confirm ONLY what cannot be reversed.
554> - `UI-5.4` Motion clarifies cause and effect: 150–250ms, ease-out, honours reduced-motion.
555> - `UI-5.5` Keep the visible focus ring. Restyle it if it clashes; NEVER remove it.
556>
557> **6.0 Navigation & IA**
558>
559> - `UI-6.1` Every screen answers: where am I, what's here, where next.
560> - `UI-6.2` Max 5 top-level destinations. Depth beats a crowded first level.
561> - `UI-6.3` Label by the user's word, not the internal one. Team vocabulary is not a taxonomy.
562> - `UI-6.4` Every state deserves a URL or a back path. Deep links and hardware back must land somewhere sensible.
563>
564> **7.0 Forms & Input**
565>
566> - `UI-7.1` Ask for less. Every field needs a reason it exists today.
567> - `UI-7.2` Labels stay visible. Placeholders are hints, NEVER labels.
568> - `UI-7.3` Validate on blur, not on keystroke. Errors sit next to the field and say how to fix it.
569> - `UI-7.4` Match keyboard to data type — correct input type, autocomplete and autocapitalise on every field.
570> - `UI-7.5` NEVER lose entered data. Preserve input across errors, navigation and refresh.
571>
572> **8.0 Mobile & Touch** _(mobile/touch surfaces)_
573>
574> - `UI-8.1` Hit target ≥44×44pt, 8px apart. The target may exceed the visible icon.
575> - `UI-8.2` Primary actions in the bottom third — that is where the thumb lives.
576> - `UI-8.3` Gestures are shortcuts, NEVER the only route. Anything swipeable is also tappable.
577> - `UI-8.4` Respect safe areas and the keyboard. Notch, home indicator and on-screen keyboard all steal space.
578>
579> **9.0 Speed & Perceived Speed**
580>
581> - `UI-9.1` Show structure before data — skeletons for known layouts, spinners only for unknown waits.
582> - `UI-9.2` Assume success optimistically. Update UI first, reconcile after, roll back visibly on failure.
583> - `UI-9.3` Reserve space for anything that loads. Images, ads and fonts must NEVER shift the layout.
584> - `UI-9.4` Design for the slow connection. Offline, timeout and retry are states, NOT edge cases.
585>
586> **Apply by role** — the clauses are one set; what you DO with them depends on the task:
587>
588> | Role | Obligation |
589> | ----------------------------------- | ----------------------------------------------------------------------------------------------------------------------------------------------------------------------- |
590> | 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`) |
591> | 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`) |
592> | 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 |
593> | SCAN / document a UI system | Record where the PROJECT deliberately deviates, so the overriding doc becomes the recorded authority |
594>
595> **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.
596
597<!-- /SYNC:ui-ux-design-principles -->
598
599<!-- SYNC:ui-wireframe-protocol -->
600
601> **UI Wireframe Protocol** — Wireframe-to-implementation flow: (1) Process design input (Figma/screenshot/sketch via visual analysis tooling). (2) Create ASCII wireframe with box-drawing chars. (3) Build component inventory with tier classification (Common/Domain-Shared/Page). (4) Document states (Default/Loading/Empty/Error). (5) Map to design tokens. (6) Define responsive breakpoints. Search existing component libraries before creating new. Progressive detail by skill level (idea=sketch, story=full tree+specs).
602
603<!-- /SYNC:ui-wireframe-protocol -->
604
605<!-- SYNC:critical-thinking-mindset -->
606
607> **Critical Thinking Mindset** — Apply critical thinking, sequential thinking. Every claim needs traced proof, confidence >80% to act.
608> **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.
609
610<!-- /SYNC:critical-thinking-mindset -->
611
612<!-- SYNC:ui-system-context:reminder -->
613
614- **MANDATORY IMPORTANT MUST ATTENTION** read frontend-patterns-reference, scss-styling-guide, design-system/README before any UI change.
615 <!-- /SYNC:ui-system-context:reminder -->
616
617<!-- SYNC:ui-ux-design-principles:reminder -->
618
619- **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.
620
621<!-- /SYNC:ui-ux-design-principles:reminder -->
622
623<!-- SYNC:ui-wireframe-protocol:reminder -->
624
625**IMPORTANT MUST ATTENTION** follow wireframe protocol: ASCII wireframe, component inventory with tiers, states table, design tokens, responsive breakpoints.
626
627<!-- /SYNC:ui-wireframe-protocol:reminder -->
628
629<!-- SYNC:ui-intent-layer:reminder -->
630
631- **MANDATORY** For UI-bearing specs, author/maintain the tech-agnostic interaction-surface layer (View Inventory + Navigation Map + observable UI States + per-story `US-/OP-/BR-`-traced flow); keep deep visual fidelity in the linked `design-spec`/mockup recorded in frontmatter; name ZERO frameworks/routes/CSS/component classes; skip ONLY for backend-only features with a stated reason.
632
633<!-- /SYNC:ui-intent-layer:reminder -->
634
635<!-- SYNC:critical-thinking-mindset:reminder -->
636
637**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.
638
639<!-- /SYNC:critical-thinking-mindset:reminder -->
640
641<!-- SYNC:ai-mistake-prevention:reminder -->
642
643**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.
644
645<!-- /SYNC:ai-mistake-prevention:reminder -->
646
647<!-- PROMPT-ENHANCE:STEP-TASK-CLOSING:START -->
648
649## Prompt-Enhance Closing Anchors
650
651**IMPORTANT MUST ATTENTION** follow declared step order for this skill; NEVER skip, reorder, or merge steps without explicit user approval
652**IMPORTANT MUST ATTENTION** for every step/sub-skill call: set `in_progress` before execution, set `completed` after execution
653**IMPORTANT MUST ATTENTION** every skipped step MUST include explicit reason; every completed step MUST include concise evidence
654**IMPORTANT MUST ATTENTION** if Task tools unavailable, maintain an equivalent step-by-step plan tracker with synchronized statuses
655
656<!-- PROMPT-ENHANCE:STEP-TASK-CLOSING:END -->
657
658<!-- SYNC:project-protocol-overlay -->
659
660> **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`.
661>
662> 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.
663
664<!-- /SYNC:project-protocol-overlay -->
665
666<!-- SYNC:project-protocol-overlay:reminder -->
667
668**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.
669
670<!-- /SYNC:project-protocol-overlay:reminder -->
671
672## Closing Reminders
673
674**IMPORTANT MUST ATTENTION Goal:** Produce a structured, tech-agnostic UI/UX design specification from a requirement/PBI/story/wireframe so a developer can rebuild the UI on ANY stack — every component, state, design token, responsive rule, and accessibility need documented and linked back to the governing Feature Spec.
675
676**IMPORTANT MUST ATTENTION main steps (do not forget any):**
677
678- **Step 0–0b first** — inventory existing related UI + map connected flows (record in §1); seed from the governing Feature Spec's §6 and reuse its view/state vocabulary verbatim — why: reinventing what exists or renaming spec terms breaks the navigable hub.
679- **Step 1–2** — route by input type (Figma→`$figma-design`, image→visual analysis, wireframe→`--mode=wireframe`, PBI/text→extract), then set Quick (§1–4) vs Full (§1–7) complexity.
680- **Step 3–6** — author component inventory (new vs existing), all 7 states + interactions, design tokens (reuse design-system), responsive breakpoints.
681- **Step 7–8** — save to `team-artifacts/design-specs/`, then set the governing Feature Spec's `design_spec:`/`mockup:` frontmatter to the saved path (frontmatter only) — why: the link-back gate fails without it.
682- **M1–M5 gate (BLOCKING)** — business-level component names, no code-prop refs in prose, cross-ref behavior by logical ID (`OP-`/`BR-`/`FR-`), one testable interpretation per state, rebuildable from spec alone — why: a tech-leaking spec cannot be rebuilt on another stack.
683
684**Protocols in force (concise digest of the SYNC/shared blocks this skill carries):**
685
686- **AI Mistake Prevention:** verify generated content against evidence, trace downstream references, verify all affected outputs, re-read after context loss, surface ambiguity.
687- **UI System Context:** read frontend-patterns, scss-styling, design-system before any UI change.
688- **UI Wireframe:** ASCII wireframe, tiered component inventory, states table, tokens, responsive breakpoints.
689- **Critical Thinking:** traced `file:line` proof per claim, confidence >80% to act.
690
691- **MANDATORY IMPORTANT MUST ATTENTION** break work into small todo tasks using task tracking BEFORE starting
692- **MANDATORY IMPORTANT MUST ATTENTION** search codebase for 3+ similar patterns before creating new code
693- **MANDATORY IMPORTANT MUST ATTENTION** cite `file:line` evidence for every claim (confidence >80% to act)
694- **MANDATORY IMPORTANT MUST ATTENTION** add a final review todo task to verify work quality
695- **MANDATORY IMPORTANT MUST ATTENTION** carry the 40 UI/UX Design Principles (`UI-1.1`–`UI-9.4`) into the spec sections so implementers INHERIT them: §5 authors empty/loading/error FIRST (`UI-1.5`) and enumerates all 5 interaction states per interactive element (`UI-5.2`); §4 DECLARES the 6-step type scale (`UI-2.5`) and the ONE 4/8px spacing unit (`UI-4.1`) instead of one-off values, with contrast targets measured and stated (4.5:1 text / 3:1 edges, `UI-3.1`); §6 keeps the focus ring visible (`UI-5.5`), colour never alone (`UI-3.3`), and ≥44×44pt targets + bottom-third primaries on touch surfaces (`UI-8.1`, `UI-8.2`); §7 records every deliberate deviation — why: an unstated clause is an unbuilt clause once the spec leaves this skill
696
697**Anti-Rationalization:**
698
699| Evasion | Rebuttal |
700| -------------------------------------------------- | --------------------------------------------------------------------------------------------------- |
701| "I'll skip the existing-UI inventory and design" | Step 0 is BLOCKING — inventory existing screens/flows first or the spec contradicts the live UI. |
702| "Naming the framework component is clearer" | M1/M2 FAIL — name by UX role; framework/CSS names live only in Evidence/frontmatter/Mermaid. |
703| "The spec already names views — I'll re-partition" | Reuse §6 view + state vocabulary verbatim; renaming breaks the navigable spec↔design hub. |
704| "Saved the file — done" | Step 8 link-back is required; set the spec's `design_spec:` frontmatter or `artifact-review` fails. |
705| "'Show feedback' describes the error state" | M4 FAIL — every state needs one observable completion marker, not a vague phrase. |
706
707**[TASK-PLANNING]** Before acting, analyze task scope and systematically break it into small todo tasks and sub-tasks using task tracking.
708
709> **[IMPORTANT]** Analyze how big the task is and break it into many small todo tasks systematically before starting — this is very important.
710
711<!-- CODEX:SYNC-PROMPT-PROTOCOLS:START -->
712
713## Hookless Prompt Protocol Mirror (Auto-Synced)
714
715Source: `.claude/.ck.json` + `.claude/skills/shared/sync-inline-versions.md` (`:full` blocks) + `.claude/scripts/lib/hookless-prompt-protocol.cjs`
716
717## [WORKFLOW-EXECUTION-PROTOCOL] [BLOCKING] Workflow Execution Protocol — MANDATORY IMPORTANT MUST CRITICAL. Do not skip for any reason.
718
719**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.
720
7211. **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.
7222. **ANALYZE:** Choose the best option: execute directly, invoke a skill, activate a standard workflow, or compose a custom step combination.
7233. **AUTO-SELECT:** Pick the best option yourself. Do not ask the user to choose between direct execution, skill, standard workflow, or custom workflow.
7244. **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.
7255. **CREATE TASKS:** task tracking for ALL workflow/skill/custom steps before execution when the selected path has multiple steps.
7266. **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.
7277. **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)
728
729## Shared AI-SDD Protocol Markers
730
731Source: `.claude/skills/shared/sync-inline-versions.md`
732
733## SYNC:ai-sdd-artifact-contract
734
735> **AI-SDD Artifact Contract** — Shared spec-driven development rules stay portable and source-owned.
736>
737> 1. Keep reusable AI-SDD principles in `.claude`; put repository-specific paths, commands, owners, products, and formats in project config/reference docs.
738> 2. Preserve cycle: `spec -> plan -> tasks -> implement -> verify -> update spec/docs`.
739> 3. Trace every requirement or invariant through decision, task, TC/test, source evidence, and docs/spec update.
740> 4. Treat code-to-spec extraction as reference-only until accepted by the canonical spec owner.
741> 5. Any supported AI tool may plan, implement, review, or verify with synced context; using multiple tools is optional.
742> 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
743> 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.
744>
745> **Active reference:** `shared/sdd-artifact-contract.md` in the active skills root.
746
747---
748
749## SYNC:ai-sdd-artifact-contract:reminder
750
751- **MANDATORY** Apply `shared/sdd-artifact-contract.md`; keep reusable AI-SDD in `.claude` and local rules in project docs.
752- **MANDATORY** Code-to-spec extraction is reference-only until canonical acceptance; any supported AI tool may execute with synced context.
753- **MANDATORY** Update `.claude` source before syncing generated mirrors; do not manually edit `.agents`, `.codex`, or `AGENTS.md`.
754- **MANDATORY** Missing or stale project config, root instruction files, or required reference docs route project-specific work through `$project-init` or the narrow setup route automatically.
755 **[TASK-PLANNING] [MANDATORY]** BEFORE executing any workflow or skill step, create/update task tracking for all planned steps, then keep it synchronized as each step starts/completes.
756
757## [LESSON-LEARNED-REMINDER] [BLOCKING] Task Planning & Continuous Improvement — MANDATORY. Do not skip.
758
759Break work into small tasks (task tracking) before starting. Add final task: "Analyze AI mistakes & lessons learned".
760
761**Extract lessons — ROOT CAUSE ONLY, not symptom fixes:**
762
7631. Name the FAILURE MODE (reasoning/assumption failure), not symptom — "assumed API existed without reading source" not "used wrong enum value".
7642. Generality test: does this failure mode apply to ≥3 contexts/codebases? If not, abstract one level up.
7653. Write as a universal rule — strip project-specific names/paths/classes. Useful on any codebase.
7664. Consolidate: multiple mistakes sharing one failure mode → ONE lesson.
7675. **Recurrence gate:** "Would this recur in future session WITHOUT this reminder?" — No → skip `$learn`.
7686. **Auto-fix gate:** "Could `$code-review`/`$code-simplifier`/`$security-review`/`$lint` catch this?" — Yes → improve review skill instead.
7697. BOTH gates pass → ask user to run `$learn`.
770 **[CRITICAL-THINKING-MINDSET]** Apply critical thinking, sequential thinking. Every claim needs traced proof, confidence >80% to act.
771 **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.
772 **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.
773 **Goal-driven execution:** Define success criteria first, loop until verified, and stop only when observable checks pass.
774 **Tests verify intent:** Tests must protect business rules/invariants and fail when the protected intent breaks, not only mirror current behavior.
775
776## Common AI Mistake Prevention (System Lessons)
777
778- **Re-read files after context compaction.** Edit requires prior Read in same context; compaction wipes read state. Re-read before editing.
779- **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.
780- **Check downstream references before deleting.** Deletions cascade doc/code staleness. Map referencing files before removal.
781- **After memory loss, check existing state before creating new.** Compaction wipes prior-work memory. Query current state to resume — never blindly duplicate.
782- **Verify AI-generated content against actual code.** AI hallucinates APIs, class names, method signatures. Grep to confirm existence before documenting/referencing.
783- **Trace full dependency chain after edits.** Changing a definition misses downstream consumers. Trace the full chain.
784- **When renaming, grep ALL consumer file types.** Some file types silently ignore missing refs (no compile error). Search code, templates, configs, generated files.
785- **Trace ALL code paths when verifying correctness.** Code existing ≠ code executing. Trace early exits, error branches, conditional skips — not just happy path.
786- **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.
787- **Verify sub-agent results after context recovery.** Background agents may finish while parent compacted — grep-verify output, don't trust assumed completion.
788- **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.
789- **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).
790- **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.
791- **When debugging, ask "whose responsibility?" before fixing.** Trace caller (wrong data) vs callee (wrong handling). Fix at responsible layer — never patch symptom site.
792- **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.
793- **Grep ALL removed names after extraction/refactoring.** Primary file "done" ≠ secondary files clean. Grep entire scope for every removed symbol before declaring complete.
794- **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.
795- **Verify ALL affected outputs, not just the first.** One build green ≠ all green. Multi-stack changes (backend/frontend/tests/docs) require verifying EVERY output.
796- **Evaluate fit before copying a nearby pattern.** Closest example ≠ matching preconditions — verify the new context shares the same constraints, base classes, scope, lifetime.
797- **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.
798- **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.
799- **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.
800- **[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.
801- **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".
802- **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.
803- **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.
804- **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.
805- **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.
806- **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.
807
808<!-- CODEX:SYNC-PROMPT-PROTOCOLS:END -->