[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"skill:01M73E4AYPRSC957J5H3X0BQNF":3,"skill-same:01M73E4AYPRSC957J5H3X0BQNF":79},{"skillNo":4,"name":5,"nameZh":6,"description":7,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":9,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":6,"sourceHtmlUrl":15,"categories":16,"keywords":17,"publishedAt":18,"linkOnly":19,"markdownContent":20,"markdownBody":21,"frontMatter":22,"refName":57,"commitSha":58,"skillDirectory":59,"forksCount":60,"repositoryHtmlUrl":61,"repositoryHomepageUrl":6,"ownerHtmlUrl":62,"ownerBlogUrl":6,"ownerDescription":63,"updatedAt":64,"install":65,"resources":78},"01M73E4AYPRSC957J5H3X0BQNF","design-review",null,"Reviews an IMPLEMENTED UI (prototype or shipped screens) against the project's design SSOT and user journeys, and drives remediation of the issues found. Covers: design-system conformance (philosophy, principles, guidelines, color system, standard icons, standard components, standard screens), interaction affordances in tables and data grids (pressable items that do nothing, or actions with no affordance), user-journey walkthroughs, and an all-agent UI\u002FUX critique session with a shared evidence-based issue log and a gated remediation loop. Use when: reviewing an implemented UI for design conformance, running a UI\u002FUX critique across the agent roster, auditing table\u002Fdata-grid interaction affordances, or walking an implementation through user journeys.","5throck\u002Fai-workspace-standards","5throck","https:\u002F\u002Favatars.githubusercontent.com\u002Fu\u002F10958964?v=4","01M73BFEH0V0VS483TW4JD49JW","01M73BFEJ5VQZJHJSFTVPAZGDM",25,"AGPL-3.0","https:\u002F\u002Fgithub.com\u002F5throck\u002Fai-workspace-standards\u002Fblob\u002Fmain\u002Ftemplates\u002Fco-design\u002F.codex\u002Fskills\u002Fdesign-review\u002FSKILL.md",[],[],"2026-11-11 15:11:14",false,"---\nname: design-review\ndescription: >\n  Reviews an IMPLEMENTED UI (prototype or shipped screens) against the project's\n  design SSOT and user journeys, and drives remediation of the issues found.\n  Covers: design-system conformance (philosophy, principles, guidelines, color\n  system, standard icons, standard components, standard screens), interaction\n  affordances in tables and data grids (pressable items that do nothing, or\n  actions with no affordance), user-journey walkthroughs, and an all-agent UI\u002FUX\n  critique session with a shared evidence-based issue log and a gated\n  remediation loop.\n  Use when: reviewing an implemented UI for design conformance, running a\n  UI\u002FUX critique across the agent roster, auditing table\u002Fdata-grid interaction\n  affordances, or walking an implementation through user journeys.\nversion: 1.1.0\nscope: co-design\nstatus: active\nowner: pm\nlast_reviewed: 2026-11-02\nprerequisites: runnable artifact under review (URL or HTML\u002Fcode files); browser automation (browser-use:control-browser or web-gui-tester) strongly recommended for Lens B\u002FC — without it, interaction findings must be labeled static-analysis\nrelates_to:\n  - skill: design-foundation\n    type: follows\n  - skill: ui-ux-design-intelligence\n    type: composes_with\n  - skill: token-usage-lint\n    type: composes_with\n  - skill: accessibility-audit\n    type: composes_with\n  - skill: service-design\n    type: composes_with\n  - skill: meeting-facilitation\n    type: composes_with\n  - skill: project-review\n    type: composes_with\nmetadata:\n  type: process\n  triggers:\n    - design review\n    - design conformance\n    - design critique\n    - ui review\n    - ux review\n    - interaction audit\n    - journey walkthrough\n---\n\n# 🔍 Skill: design-review\n\n## Context\n\nClose the loop that the creation skills open. `design-foundation`,\n`ui-ux-design-intelligence`, and `service-design` produce design systems,\ncomponents, and journeys; this skill verifies that what was actually\n**implemented** matches them, finds what does not, and routes fixes through the\nPM Gateway. One skill, one issue log, one remediation loop — across three\nreview lenses and one all-agent critique.\n\n> **Provenance**: v1.1.0 absorbs the field-validated practices of the\n> 2026-10-22 co-newbiz S1–S12 commission run (that project's\n> `docs\u002Freports\u002F2026-10-22-design-journey-review.md`): machine-baseline-first,\n> in-review baseline authoring, entry-pattern consolidation, multi-round\n> critique with per-surface verdicts, standards fold-back, and the machine\n> ratchet.\n\n## When to Use\n\n**Use when:**\n- \"Review the prototype \u002F screens against our design system\"\n- \"Check whether the tables' pressable items actually work\" (interaction-affordance audit)\n- \"Walk the implementation through the user journeys and find problems\"\n- \"Run a design critique panel with all agents\" — via `meeting-facilitation`\n- Post-fix verification after a previous design review\n\n**Do NOT use when:**\n- Creating or extending the design system itself → `design-foundation`\n- An accessibility-only audit → `accessibility-audit` (this skill *invokes* it as one machine check)\n- A project-wide governance\u002Fstructure review → `project-review`\n\n## Input Contracts (fail-fast per lens)\n\nEach lens declares its own inputs. A missing input is **logged as a process\nfinding** (\"no conformance baseline declared\") and that lens is skipped — the\nrun never silently degrades to guesses.\n\n| Lens | Required input | Where it lives |\n|------|----------------|----------------|\n| A — Conformance | Design SSOT: `docs\u002Fdesign.md` `design_decisions`, token implementation, icon vocabulary, component inventory, declared screen patterns | `design-foundation` outputs |\n| B — Interaction | Runnable artifact: URL or HTML\u002Fcode files that can be rendered and clicked | prototype-engineer output \u002F app build |\n| C — Journey | Journey map or PRD user flows defining personas, tasks, success criteria | `service-design` output \u002F PRD |\n\n## Scope Modes\n\n| Mode | Coverage | When |\n|------|----------|------|\n| `full` | All lenses, all screens, full critique | Initial review, major milestone |\n| `scoped \u003Clens or screens>` | Named lenses\u002Fscreens only | Post-fix verification, targeted concern |\n| `sampled` | Stratified screen sample — every component, pattern, and journey stage represented at least once | Large prototypes; budget-constrained runs |\n\nPick the smallest mode covering the request and state the choice in the report.\n\n## Execution Steps\n\n### Step 0 — Pre-flight\n\n1. Confirm the review target (paths\u002FURL) and resolve the mode.\n2. **Run the machine baseline FIRST** (`project-review` Step 0 pattern) — the design-lint battery plus `token-usage-lint` and, where runnable, `accessibility-audit` — before any agent dispatch. Record PASS\u002FFAIL\u002FSKIP per check. A gate that silently SKIPS (e.g. design-lint with no `designLint.scanRoots` configured in `workspace-schema.json`) is itself a baseline finding: activating the gate is a review outcome (machine ratchet, Step 6).\n3. Check each lens's input contract (table above). Record skip\u002Fgap decisions. A missing baseline is a process finding; with user approval the review may **author the missing artifact in-session** instead of skipping the lens — the 2026-10-22 co-newbiz run authored a 42-entry icon-vocabulary decision-of-record during the review rather than dropping the icons check.\n4. Build the screen inventory; for `sampled`, select the sample and justify it.\n\n### Step 1 — Lens A: Design-System Conformance Audit\n\nAudit the rendered screens against the SSOT, in this order:\n\n1. **Principles** — do implemented screens contradict any numbered principle in `docs\u002Fdesign.md`?\n2. **Tokens** — no hard-coded values in governed components (run `token-usage-lint`); no semantic-layer bypass; default and alternate themes both render correctly.\n3. **Color system** — status colors resolve from semantic tokens; status is multi-encoded (color + icon + text), never color alone.\n4. **Icons** — every icon comes from the registered icon vocabulary; consistent size\u002Fstroke via icon tokens; every icon-only control carries `aria-label`.\n5. **Components** — UI uses declared inventory components only (no ad-hoc lookalikes); all states present (default, hover, focus, active, disabled, error).\n6. **Standard screens\u002Fpatterns** — declared patterns are followed where applicable; deviations are recorded either as findings or as candidate new decisions (never ignored).\n\n**Method (mandatory mix):**\n- Machine checks: `token-usage-lint`, `accessibility-audit` (axe-core, WCAG 2.1 AA).\n- Rendered inspection: render each screen to PNG and inspect it. Judging Lens A\n  findings from source code alone is prohibited — token drift, contrast, and\n  layout issues only show in the render.\n- Dispatch read-only audits in parallel where available (cap: 4 concurrent agents).\n\n### Step 2 — Lens B: Interaction-Affordance Audit (Tables & Data Grids)\n\nVerify, by actually clicking in a browser, that affordance and behavior agree:\n\n| Check | Rule |\n|-------|------|\n| Affordance → behavior | Everything that *looks* pressable (pointer cursor, link styling, chevron, button look) performs a real action; console stays clean |\n| Behavior → affordance | Every intended action has a visible affordance — no invisible\u002Fdead click targets, no click-swallowing overlays |\n| Row vs cell | Row-click behavior consistent; nested controls (checkbox, link, menu) do not conflict with the row target |\n| Sort headers | Clickable, order toggles, sort indicator updates |\n| Pagination & selection | Page links\u002Fnext-prev work; select-all and row selection stay consistent |\n| States | hover\u002Ffocus\u002Factive\u002Fdisabled present and distinguishable; loading, empty, and error states defined for the grid |\n| Keyboard | Interactive cells reachable by tab; Enter activates; focus visible |\n| Entry\u002Faction pattern consolidation | Each user action has ONE sanctioned entry pattern (entry = label-cell link; row actions = dedicated last column; gated saves = focusable `aria-disabled` + hint; deletes = row-specific `aria-label`). Coexisting entry patterns for the same action is a finding even when every affordance works — the 2026-10-22 co-newbiz run found 0 false affordances but 6 coexisting entry patterns for one action. Structural table standards (e.g. list-table: fixed layout, width cap, every column headed including the `sr-only` actions header, date-column width class) are checked against their declared rules; `design-lint`'s `list-table` sub-check enforces this mechanically where configured |\n\n**Method:** drive the artifact with `browser-use:control-browser` or\n`web-gui-tester`. Capture before\u002Fafter screenshots and console output per\nfinding. If no browser tool is available in the session, run static\nDOM\u002Fcode analysis only and **label every Lens B finding\n`static-analysis (interaction unverified)`** — do not present it as verified.\n\n### Step 3 — Lens C: Journey Walkthrough\n\n1. Take the journey map (or PRD user flows) and list its tasks with success criteria.\n2. For each persona\u002Ftask, walk the implementation end-to-end: entry point → each step → outcome → exit. Record a step-by-step trace with screenshots.\n3. **Diversity-profile walks (ADR-0068, mandatory)** — repeat each mapped journey once per profile: aging, cognitive, situational, motor-constrained. WCAG conformance does not substitute for this step.\n4. Classify journey issues: dead end (no forward path), missing state (loading\u002Fempty\u002Ferror), silent action (no feedback), broken error recovery, flow mismatch vs the map, unnecessary-step friction.\n\n### Step 4 — All-Agent Design Critique Panel\n\nDelegate to `meeting-facilitation` with topic \"design review critique\" and the\nreview findings from Steps 1–3 as the agenda input. Attach this lens map:\n\n| Agent | Critique lens |\n|-------|---------------|\n| design-lead | Overall direction, cross-screen consistency, standard-screen adherence |\n| visual-designer | Tokens, color system, iconography, component conformance |\n| typography-expert | Type hierarchy, readability, tabular numerals in data UI |\n| ux-researcher | Usability heuristics, cognitive load, error recovery |\n| service-designer | Journey integrity, touchpoint continuity, diversity-profile gaps |\n| prototype-engineer | Interaction quality, state completeness, feasibility of fixes |\n| storyteller | Brand-narrative consistency, principles adherence |\n| pm | Facilitates; synthesizes; guarantees the dissent seat |\n\n**Rubric** — each dimension scored 1–5 (1 = fails, 3 = acceptable with noted\ndebt, 5 = exemplary) with one evidence-backed sentence per score: visual\nconsistency, typography, usability, journey integrity, interaction quality,\naccessibility, brand narrative.\n\n**Governance invariants (inherited from `meeting-facilitation`):** a designated\ndissent seat challenges the emerging consensus; dissent is recorded verbatim;\nthe synthesized outcome is a **PROPOSAL for the user** — the critique never\ndecides remediation itself. Transcript → `memory\u002Fmeeting-YYYY-MM-DD-design-review.md`.\n\n**Cost guidance:** the full roster critique is expensive — reserve it for\n`full` mode; `scoped` runs invite only the agents owning the affected lenses.\n\n**Round structure (field-validated by the 2026-10-22 co-newbiz S1–S12 run):**\ntwo rounds beat one — Round 1 applies the technical + accessibility lenses to\nthe just-fixed surfaces (regressions get caught while still cheap), Round 2 is\nthe full judge panel over the whole surface set. Verdicts use\nPASS \u002F PASS-with-notes \u002F FAIL per surface per lens; open notes route to\nimmediate fixes or a ticket bundle. Close the critique by naming the\n**top-3 improvements of record** — the changes that most moved the product —\nso the panel's judgment is preserved, not just its findings.\n\n### Step 5 — Unified Issue Log\n\n1. Merge findings from all lenses and the critique. Deduplicate by root cause\n   (credit all discoverers; keep the most specific evidence and the highest severity).\n2. Persist the report to `docs\u002Freports\u002FYYYY-MM-DD-design-review-\u003Cscope>.md`:\n\n```markdown\n## Findings\n\n### 🔴 Critical — blocks use or breaks the design system\n| # | Issue | Lens | Evidence | Fix owner | Status |\n|---|-------|------|----------|-----------|--------|\n\n### 🟡 High — fix before next milestone   \u003C!-- same columns -->\n### 🟢 Moderate — fix within the cycle    \u003C!-- same columns -->\n### ℹ️ Low \u002F improvements\n### ✅ Strengths\n```\n\n3. Every finding row **must** carry evidence: screenshot path, console output,\n   or `file:line`, plus a repro step. A finding without evidence is rejected at dedup.\n4. Append the critique rubric table and the meeting transcript link to the report.\n\n### Step 6 — Remediation Proposal & Gate\n\n1. Present the findings as a **remediation proposal** to the user (severity,\n   effort, owner per finding). Fixes execute only on user approval — the\n   review itself decides nothing.\n2. On approval, dispatch through the PM Gateway to the role-boundary owner:\n   `visual-designer` (tokens\u002Fcolor\u002Ficons), `typography-expert` (type),\n   `prototype-engineer` (interaction\u002Fmarkup\u002Fbehavior), `service-designer`\n   (journey-level redesign — re-enters the design phase).\n3. **Design Gate (ADR-0074)**: every code change carries spec activity —\n   `docs\u002Fdesigns\u002F\u003Cspec-id>-design.md` registered via\n   `bun scripts\u002Fspec-register.ts` before `\u002Fsync`. Trivial fixes use the\n   ADR-0074 exemption codes.\n4. **Standards fold-back**: when a finding reveals the STANDARD itself was\n   wrong, stale, or silent — not the implementation — route the fix to the\n   design SSOT documents rather than the code: record a dated\n   decision-of-record annotation (registered pattern variant, deprecated alias\n   with migration note, standard anatomy rule) in the pattern \u002F component \u002F\n   tokens docs. The 2026-10-22 co-newbiz run registered an S11 tab-bar pattern\n   variant, deprecated the `.btn-premium` alias, and standardized the\n   head-action anatomy and AI-button format this way. Standards updates pass\n   the same Design Gate as code.\n5. **Machine ratchet**: every finding class a linter could catch (raw token\n   values, off-inventory components, unregistered icons, list-table rule\n   violations) ends as a lint-hardening ticket or a gate activation so the\n   next review catches it mechanically — found-by-agent ↓, caught-by-script ↑\n   (the `project-review` ratchet principle). Framework-level gaps discovered\n   during a project review are filed to the workspace (L0) backlog.\n6. **Re-verify**: re-run the affected lenses on the changed screens only\n   (`scoped` mode) and append a `## Verification` section to the report with\n   closure status per finding.\n7. **Deferred items**: follow the `project-review` Step 5 ticket pattern at\n   the workspace root's ticket CLI (L0-only tooling, not shipped to variant\n   projects); in variant projects, record them in the report and the session\n   memory log.\n\n## Output Format\n\n- **Review report**: `docs\u002Freports\u002FYYYY-MM-DD-design-review-\u003Cscope>.md` — mode,\n  lens skip log, findings tables with evidence, rubric scores, verification results\n- **Critique transcript**: `memory\u002Fmeeting-YYYY-MM-DD-design-review.md`\n- **Session log entry**: `memory\u002FYYYY-MM-DD.md` (four mandatory sections)\n- **Remediation specs**: `docs\u002Fdesigns\u002F\u003Cspec-id>-design.md` per ADR-0074 (code changes only)\n\n## Boundaries\n\n- Does **not** create or extend design systems → `design-foundation` (this skill follows it and audits against its outputs).\n- Does **not** replace an accessibility-only audit → invokes `accessibility-audit` as one Lens A machine check.\n- Does **not** review project structure, scripts, or governance → `project-review`.\n- The critique phase **proposes**, never decides (ADR-0061\u002F`meeting-facilitation` invariants).\n\n## Related Skills\n\n- **design-foundation** — produces the conformance baseline this skill audits against\n- **token-usage-lint** — machine check for hard-coded values (Lens A)\n- **accessibility-audit** — axe-core WCAG 2.1 AA machine check (Lens A)\n- **service-design** — journey maps consumed by Lens C\n- **meeting-facilitation** — runs the all-agent critique (Step 4)\n- **project-review** — ticket pattern for deferred findings; project-wide reviews\n","# 🔍 Skill: design-review\n\n## Context\n\nClose the loop that the creation skills open. `design-foundation`,\n`ui-ux-design-intelligence`, and `service-design` produce design systems,\ncomponents, and journeys; this skill verifies that what was actually\n**implemented** matches them, finds what does not, and routes fixes through the\nPM Gateway. One skill, one issue log, one remediation loop — across three\nreview lenses and one all-agent critique.\n\n> **Provenance**: v1.1.0 absorbs the field-validated practices of the\n> 2026-10-22 co-newbiz S1–S12 commission run (that project's\n> `docs\u002Freports\u002F2026-10-22-design-journey-review.md`): machine-baseline-first,\n> in-review baseline authoring, entry-pattern consolidation, multi-round\n> critique with per-surface verdicts, standards fold-back, and the machine\n> ratchet.\n\n## When to Use\n\n**Use when:**\n- \"Review the prototype \u002F screens against our design system\"\n- \"Check whether the tables' pressable items actually work\" (interaction-affordance audit)\n- \"Walk the implementation through the user journeys and find problems\"\n- \"Run a design critique panel with all agents\" — via `meeting-facilitation`\n- Post-fix verification after a previous design review\n\n**Do NOT use when:**\n- Creating or extending the design system itself → `design-foundation`\n- An accessibility-only audit → `accessibility-audit` (this skill *invokes* it as one machine check)\n- A project-wide governance\u002Fstructure review → `project-review`\n\n## Input Contracts (fail-fast per lens)\n\nEach lens declares its own inputs. A missing input is **logged as a process\nfinding** (\"no conformance baseline declared\") and that lens is skipped — the\nrun never silently degrades to guesses.\n\n| Lens | Required input | Where it lives |\n|------|----------------|----------------|\n| A — Conformance | Design SSOT: `docs\u002Fdesign.md` `design_decisions`, token implementation, icon vocabulary, component inventory, declared screen patterns | `design-foundation` outputs |\n| B — Interaction | Runnable artifact: URL or HTML\u002Fcode files that can be rendered and clicked | prototype-engineer output \u002F app build |\n| C — Journey | Journey map or PRD user flows defining personas, tasks, success criteria | `service-design` output \u002F PRD |\n\n## Scope Modes\n\n| Mode | Coverage | When |\n|------|----------|------|\n| `full` | All lenses, all screens, full critique | Initial review, major milestone |\n| `scoped \u003Clens or screens>` | Named lenses\u002Fscreens only | Post-fix verification, targeted concern |\n| `sampled` | Stratified screen sample — every component, pattern, and journey stage represented at least once | Large prototypes; budget-constrained runs |\n\nPick the smallest mode covering the request and state the choice in the report.\n\n## Execution Steps\n\n### Step 0 — Pre-flight\n\n1. Confirm the review target (paths\u002FURL) and resolve the mode.\n2. **Run the machine baseline FIRST** (`project-review` Step 0 pattern) — the design-lint battery plus `token-usage-lint` and, where runnable, `accessibility-audit` — before any agent dispatch. Record PASS\u002FFAIL\u002FSKIP per check. A gate that silently SKIPS (e.g. design-lint with no `designLint.scanRoots` configured in `workspace-schema.json`) is itself a baseline finding: activating the gate is a review outcome (machine ratchet, Step 6).\n3. Check each lens's input contract (table above). Record skip\u002Fgap decisions. A missing baseline is a process finding; with user approval the review may **author the missing artifact in-session** instead of skipping the lens — the 2026-10-22 co-newbiz run authored a 42-entry icon-vocabulary decision-of-record during the review rather than dropping the icons check.\n4. Build the screen inventory; for `sampled`, select the sample and justify it.\n\n### Step 1 — Lens A: Design-System Conformance Audit\n\nAudit the rendered screens against the SSOT, in this order:\n\n1. **Principles** — do implemented screens contradict any numbered principle in `docs\u002Fdesign.md`?\n2. **Tokens** — no hard-coded values in governed components (run `token-usage-lint`); no semantic-layer bypass; default and alternate themes both render correctly.\n3. **Color system** — status colors resolve from semantic tokens; status is multi-encoded (color + icon + text), never color alone.\n4. **Icons** — every icon comes from the registered icon vocabulary; consistent size\u002Fstroke via icon tokens; every icon-only control carries `aria-label`.\n5. **Components** — UI uses declared inventory components only (no ad-hoc lookalikes); all states present (default, hover, focus, active, disabled, error).\n6. **Standard screens\u002Fpatterns** — declared patterns are followed where applicable; deviations are recorded either as findings or as candidate new decisions (never ignored).\n\n**Method (mandatory mix):**\n- Machine checks: `token-usage-lint`, `accessibility-audit` (axe-core, WCAG 2.1 AA).\n- Rendered inspection: render each screen to PNG and inspect it. Judging Lens A\n  findings from source code alone is prohibited — token drift, contrast, and\n  layout issues only show in the render.\n- Dispatch read-only audits in parallel where available (cap: 4 concurrent agents).\n\n### Step 2 — Lens B: Interaction-Affordance Audit (Tables & Data Grids)\n\nVerify, by actually clicking in a browser, that affordance and behavior agree:\n\n| Check | Rule |\n|-------|------|\n| Affordance → behavior | Everything that *looks* pressable (pointer cursor, link styling, chevron, button look) performs a real action; console stays clean |\n| Behavior → affordance | Every intended action has a visible affordance — no invisible\u002Fdead click targets, no click-swallowing overlays |\n| Row vs cell | Row-click behavior consistent; nested controls (checkbox, link, menu) do not conflict with the row target |\n| Sort headers | Clickable, order toggles, sort indicator updates |\n| Pagination & selection | Page links\u002Fnext-prev work; select-all and row selection stay consistent |\n| States | hover\u002Ffocus\u002Factive\u002Fdisabled present and distinguishable; loading, empty, and error states defined for the grid |\n| Keyboard | Interactive cells reachable by tab; Enter activates; focus visible |\n| Entry\u002Faction pattern consolidation | Each user action has ONE sanctioned entry pattern (entry = label-cell link; row actions = dedicated last column; gated saves = focusable `aria-disabled` + hint; deletes = row-specific `aria-label`). Coexisting entry patterns for the same action is a finding even when every affordance works — the 2026-10-22 co-newbiz run found 0 false affordances but 6 coexisting entry patterns for one action. Structural table standards (e.g. list-table: fixed layout, width cap, every column headed including the `sr-only` actions header, date-column width class) are checked against their declared rules; `design-lint`'s `list-table` sub-check enforces this mechanically where configured |\n\n**Method:** drive the artifact with `browser-use:control-browser` or\n`web-gui-tester`. Capture before\u002Fafter screenshots and console output per\nfinding. If no browser tool is available in the session, run static\nDOM\u002Fcode analysis only and **label every Lens B finding\n`static-analysis (interaction unverified)`** — do not present it as verified.\n\n### Step 3 — Lens C: Journey Walkthrough\n\n1. Take the journey map (or PRD user flows) and list its tasks with success criteria.\n2. For each persona\u002Ftask, walk the implementation end-to-end: entry point → each step → outcome → exit. Record a step-by-step trace with screenshots.\n3. **Diversity-profile walks (ADR-0068, mandatory)** — repeat each mapped journey once per profile: aging, cognitive, situational, motor-constrained. WCAG conformance does not substitute for this step.\n4. Classify journey issues: dead end (no forward path), missing state (loading\u002Fempty\u002Ferror), silent action (no feedback), broken error recovery, flow mismatch vs the map, unnecessary-step friction.\n\n### Step 4 — All-Agent Design Critique Panel\n\nDelegate to `meeting-facilitation` with topic \"design review critique\" and the\nreview findings from Steps 1–3 as the agenda input. Attach this lens map:\n\n| Agent | Critique lens |\n|-------|---------------|\n| design-lead | Overall direction, cross-screen consistency, standard-screen adherence |\n| visual-designer | Tokens, color system, iconography, component conformance |\n| typography-expert | Type hierarchy, readability, tabular numerals in data UI |\n| ux-researcher | Usability heuristics, cognitive load, error recovery |\n| service-designer | Journey integrity, touchpoint continuity, diversity-profile gaps |\n| prototype-engineer | Interaction quality, state completeness, feasibility of fixes |\n| storyteller | Brand-narrative consistency, principles adherence |\n| pm | Facilitates; synthesizes; guarantees the dissent seat |\n\n**Rubric** — each dimension scored 1–5 (1 = fails, 3 = acceptable with noted\ndebt, 5 = exemplary) with one evidence-backed sentence per score: visual\nconsistency, typography, usability, journey integrity, interaction quality,\naccessibility, brand narrative.\n\n**Governance invariants (inherited from `meeting-facilitation`):** a designated\ndissent seat challenges the emerging consensus; dissent is recorded verbatim;\nthe synthesized outcome is a **PROPOSAL for the user** — the critique never\ndecides remediation itself. Transcript → `memory\u002Fmeeting-YYYY-MM-DD-design-review.md`.\n\n**Cost guidance:** the full roster critique is expensive — reserve it for\n`full` mode; `scoped` runs invite only the agents owning the affected lenses.\n\n**Round structure (field-validated by the 2026-10-22 co-newbiz S1–S12 run):**\ntwo rounds beat one — Round 1 applies the technical + accessibility lenses to\nthe just-fixed surfaces (regressions get caught while still cheap), Round 2 is\nthe full judge panel over the whole surface set. Verdicts use\nPASS \u002F PASS-with-notes \u002F FAIL per surface per lens; open notes route to\nimmediate fixes or a ticket bundle. Close the critique by naming the\n**top-3 improvements of record** — the changes that most moved the product —\nso the panel's judgment is preserved, not just its findings.\n\n### Step 5 — Unified Issue Log\n\n1. Merge findings from all lenses and the critique. Deduplicate by root cause\n   (credit all discoverers; keep the most specific evidence and the highest severity).\n2. Persist the report to `docs\u002Freports\u002FYYYY-MM-DD-design-review-\u003Cscope>.md`:\n\n```markdown\n## Findings\n\n### 🔴 Critical — blocks use or breaks the design system\n| # | Issue | Lens | Evidence | Fix owner | Status |\n|---|-------|------|----------|-----------|--------|\n\n### 🟡 High — fix before next milestone   \u003C!-- same columns -->\n### 🟢 Moderate — fix within the cycle    \u003C!-- same columns -->\n### ℹ️ Low \u002F improvements\n### ✅ Strengths\n```\n\n3. Every finding row **must** carry evidence: screenshot path, console output,\n   or `file:line`, plus a repro step. A finding without evidence is rejected at dedup.\n4. Append the critique rubric table and the meeting transcript link to the report.\n\n### Step 6 — Remediation Proposal & Gate\n\n1. Present the findings as a **remediation proposal** to the user (severity,\n   effort, owner per finding). Fixes execute only on user approval — the\n   review itself decides nothing.\n2. On approval, dispatch through the PM Gateway to the role-boundary owner:\n   `visual-designer` (tokens\u002Fcolor\u002Ficons), `typography-expert` (type),\n   `prototype-engineer` (interaction\u002Fmarkup\u002Fbehavior), `service-designer`\n   (journey-level redesign — re-enters the design phase).\n3. **Design Gate (ADR-0074)**: every code change carries spec activity —\n   `docs\u002Fdesigns\u002F\u003Cspec-id>-design.md` registered via\n   `bun scripts\u002Fspec-register.ts` before `\u002Fsync`. Trivial fixes use the\n   ADR-0074 exemption codes.\n4. **Standards fold-back**: when a finding reveals the STANDARD itself was\n   wrong, stale, or silent — not the implementation — route the fix to the\n   design SSOT documents rather than the code: record a dated\n   decision-of-record annotation (registered pattern variant, deprecated alias\n   with migration note, standard anatomy rule) in the pattern \u002F component \u002F\n   tokens docs. The 2026-10-22 co-newbiz run registered an S11 tab-bar pattern\n   variant, deprecated the `.btn-premium` alias, and standardized the\n   head-action anatomy and AI-button format this way. Standards updates pass\n   the same Design Gate as code.\n5. **Machine ratchet**: every finding class a linter could catch (raw token\n   values, off-inventory components, unregistered icons, list-table rule\n   violations) ends as a lint-hardening ticket or a gate activation so the\n   next review catches it mechanically — found-by-agent ↓, caught-by-script ↑\n   (the `project-review` ratchet principle). Framework-level gaps discovered\n   during a project review are filed to the workspace (L0) backlog.\n6. **Re-verify**: re-run the affected lenses on the changed screens only\n   (`scoped` mode) and append a `## Verification` section to the report with\n   closure status per finding.\n7. **Deferred items**: follow the `project-review` Step 5 ticket pattern at\n   the workspace root's ticket CLI (L0-only tooling, not shipped to variant\n   projects); in variant projects, record them in the report and the session\n   memory log.\n\n## Output Format\n\n- **Review report**: `docs\u002Freports\u002FYYYY-MM-DD-design-review-\u003Cscope>.md` — mode,\n  lens skip log, findings tables with evidence, rubric scores, verification results\n- **Critique transcript**: `memory\u002Fmeeting-YYYY-MM-DD-design-review.md`\n- **Session log entry**: `memory\u002FYYYY-MM-DD.md` (four mandatory sections)\n- **Remediation specs**: `docs\u002Fdesigns\u002F\u003Cspec-id>-design.md` per ADR-0074 (code changes only)\n\n## Boundaries\n\n- Does **not** create or extend design systems → `design-foundation` (this skill follows it and audits against its outputs).\n- Does **not** replace an accessibility-only audit → invokes `accessibility-audit` as one Lens A machine check.\n- Does **not** review project structure, scripts, or governance → `project-review`.\n- The critique phase **proposes**, never decides (ADR-0061\u002F`meeting-facilitation` invariants).\n\n## Related Skills\n\n- **design-foundation** — produces the conformance baseline this skill audits against\n- **token-usage-lint** — machine check for hard-coded values (Lens A)\n- **accessibility-audit** — axe-core WCAG 2.1 AA machine check (Lens A)\n- **service-design** — journey maps consumed by Lens C\n- **meeting-facilitation** — runs the all-agent critique (Step 4)\n- **project-review** — ticket pattern for deferred findings; project-wide reviews\n",{"name":5,"owner":23,"scope":24,"status":25,"version":26,"metadata":27,"relates_to":37,"description":54,"last_reviewed":55,"prerequisites":56},"pm","co-design","active","1.1.0",{"type":28,"triggers":29},"process",[30,31,32,33,34,35,36],"design review","design conformance","design critique","ui review","ux review","interaction audit","journey walkthrough",[38,41,44,46,48,50,52],{"type":39,"skill":40},"follows","design-foundation",{"type":42,"skill":43},"composes_with","ui-ux-design-intelligence",{"type":42,"skill":45},"token-usage-lint",{"type":42,"skill":47},"accessibility-audit",{"type":42,"skill":49},"service-design",{"type":42,"skill":51},"meeting-facilitation",{"type":42,"skill":53},"project-review","Reviews an IMPLEMENTED UI (prototype or shipped screens) against the project's design SSOT and user journeys, and drives remediation of the issues found. Covers: design-system conformance (philosophy, principles, guidelines, color system, standard icons, standard components, standard screens), interaction affordances in tables and data grids (pressable items that do nothing, or actions with no affordance), user-journey walkthroughs, and an all-agent UI\u002FUX critique session with a shared evidence-based issue log and a gated remediation loop. Use when: reviewing an implemented UI for design conformance, running a UI\u002FUX critique across the agent roster, auditing table\u002Fdata-grid interaction affordances, or walking an implementation through user journeys.\n",1793577600000,"runnable artifact under review (URL or HTML\u002Fcode files); browser automation (browser-use:control-browser or web-gui-tester) strongly recommended for Lens B\u002FC — without it, interaction findings must be labeled static-analysis","main","1162ccc568cf90c01b3886d6143cbfd851a37883","templates\u002Fco-design\u002F.codex\u002Fskills\u002Fdesign-review",9,"https:\u002F\u002Fgithub.com\u002F5throck\u002Fai-workspace-standards","https:\u002F\u002Fgithub.com\u002F5throck","Experienced in digital transformation, enterprise IT, startup acceleration, and business strategy. Focused on building practical systems that connect technolo","2026-11-14 21:22:34",{"source":8,"skillName":5,"commands":66,"prompt":76,"zipAvailable":77},[67,70,73],{"tool":68,"command":69},"npx","npx skills add 5throck\u002Fai-workspace-standards --skill design-review",{"tool":71,"command":72},"bunx","bunx skills add 5throck\u002Fai-workspace-standards --skill design-review",{"tool":74,"command":75},"pnpm","pnpm dlx skills add 5throck\u002Fai-workspace-standards --skill design-review","安装技能，技能地址：https:\u002F\u002Fgithub.com\u002F5throck\u002Fai-workspace-standards\u002Fblob\u002Fmain\u002Ftemplates\u002Fco-design\u002F.codex\u002Fskills\u002Fdesign-review\u002FSKILL.md",true,[],[80,88,93,98,103,108],{"skillNo":81,"name":82,"nameZh":6,"description":83,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":9,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":6,"sourceHtmlUrl":84,"categories":85,"keywords":86,"publishedAt":87},"01M7BBJDYHJMGTJYSC37JVWY8S","i18n-layout","Text layout and encoding guidance: character encoding (UTF-8, legacy Korean code pages, BOM hazards), line endings, RTL\u002Fbidi handling, and script-specific font selection (Hangul focus). Use when: encoding is corrupted or mojibake appears, an RTL locale must render correctly, fonts must be chosen for a script, or CRLF\u002FBOM issues surface.","https:\u002F\u002Fgithub.com\u002F5throck\u002Fai-workspace-standards\u002Fblob\u002Fmain\u002Ftemplates\u002Fco-work\u002Fskills\u002Fi18n-layout\u002FSKILL.md",[],[],"2026-11-15 14:19:17",{"skillNo":89,"name":82,"nameZh":6,"description":83,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":9,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":6,"sourceHtmlUrl":90,"categories":91,"keywords":92,"publishedAt":87},"01M7BBJ9XHAAS1AVGKZZ841VBZ","https:\u002F\u002Fgithub.com\u002F5throck\u002Fai-workspace-standards\u002Fblob\u002Fmain\u002Ftemplates\u002Fco-work\u002F.hermes\u002Fskills\u002Fi18n-layout\u002FSKILL.md",[],[],{"skillNo":94,"name":82,"nameZh":6,"description":83,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":9,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":6,"sourceHtmlUrl":95,"categories":96,"keywords":97,"publishedAt":87},"01M7BBJ5C3X8TKFKG5R9P9P49W","https:\u002F\u002Fgithub.com\u002F5throck\u002Fai-workspace-standards\u002Fblob\u002Fmain\u002Ftemplates\u002Fco-work\u002F.gemini\u002Fskills\u002Fi18n-layout\u002FSKILL.md",[],[],{"skillNo":99,"name":82,"nameZh":6,"description":83,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":9,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":6,"sourceHtmlUrl":100,"categories":101,"keywords":102,"publishedAt":87},"01M7BBJ1B52X0QMJ2QVBMPJB37","https:\u002F\u002Fgithub.com\u002F5throck\u002Fai-workspace-standards\u002Fblob\u002Fmain\u002Ftemplates\u002Fco-work\u002F.codex\u002Fskills\u002Fi18n-layout\u002FSKILL.md",[],[],{"skillNo":104,"name":82,"nameZh":6,"description":83,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":9,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":6,"sourceHtmlUrl":105,"categories":106,"keywords":107,"publishedAt":87},"01M7BBHX91BW2Q2R276MRX5AW3","https:\u002F\u002Fgithub.com\u002F5throck\u002Fai-workspace-standards\u002Fblob\u002Fmain\u002Ftemplates\u002Fco-work\u002F.claude\u002Fskills\u002Fi18n-layout\u002FSKILL.md",[],[],{"skillNo":109,"name":82,"nameZh":6,"description":83,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":9,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":6,"sourceHtmlUrl":110,"categories":111,"keywords":112,"publishedAt":87},"01M7BBHQWGXPSFRWN1PRC2NMMV","https:\u002F\u002Fgithub.com\u002F5throck\u002Fai-workspace-standards\u002Fblob\u002Fmain\u002Ftemplates\u002Fco-work\u002F.agents\u002Fskills\u002Fi18n-layout\u002FSKILL.md",[],[]]