From startup-assessment
Centralized visual identity and quality contract for all assessment outputs — both interactive CP artifacts (Cowork React) and final HTML/PDF reports. This is the single source of truth for colors, typography, component specs, chart palettes, and quality standards. Referenced by interactive-review and html-dashboard skills. Trigger phrases: "design system", "visual identity", "brand standards", "style guide", "quality standards".
How this skill is triggered — by the user, by Claude, or both
Slash command
/startup-assessment:design-systemThe summary Claude sees in its skill listing — used to decide when to auto-load this skill
This is the **single source of truth** for visual identity across the entire assessment workflow. Both interactive CP artifacts (rendered in Cowork) and final HTML/PDF deliverables reference this file to ensure one consistent, professional brand from first interaction to final report.
This is the single source of truth for visual identity across the entire assessment workflow. Both interactive CP artifacts (rendered in Cowork) and final HTML/PDF deliverables reference this file to ensure one consistent, professional brand from first interaction to final report.
Two rendering contexts, one visual identity:
| Context | Technology | Theme | Skill |
|---|---|---|---|
| CP Artifacts (in-conversation) | React 18, Tailwind, Shadcn, Recharts | Dark (slate-900 base) | interactive-review |
| Final Reports (standalone files) | Self-contained HTML, Chart.js | Light (white base) | html-dashboard |
The theme differs by context (dark UI for Cowork's dark conversation pane, light for standalone files optimized for reading and print). But the semantic colors, typography, component proportions, and chart palettes are identical across both.
These are absolute. Every surface that renders a determination, severity, confidence, or assessment mode must use exactly these values. No approximations.
| Determination | Hex | RGB | Usage |
|---|---|---|---|
| GO | #22c55e | 34, 197, 94 | Proceed with confidence |
| CONDITIONAL GO | #3b82f6 | 59, 130, 246 | Proceed with identified conditions |
| CONDITIONAL HOLD | #f59e0b | 245, 158, 11 | Hold pending resolution |
| NO-GO | #ef4444 | 239, 68, 68 | Do not proceed |
Backgrounds (10% opacity of determination color):
rgba(34, 197, 94, 0.1)rgba(59, 130, 246, 0.1)rgba(245, 158, 11, 0.1)rgba(239, 68, 68, 0.1)Borders (30% opacity):
rgba(34, 197, 94, 0.3)rgba(59, 130, 246, 0.3)rgba(245, 158, 11, 0.3)rgba(239, 68, 68, 0.3)| Severity | Hex | RGB |
|---|---|---|
| Critical | #ef4444 | 239, 68, 68 |
| Significant | #f97316 | 249, 115, 22 |
| Moderate | #f59e0b | 245, 158, 11 |
| Minor | #94a3b8 | 148, 163, 184 |
| Level | Hex | RGB |
|---|---|---|
| High | #22c55e | 34, 197, 94 |
| Medium | #f59e0b | 245, 158, 11 |
| Low | #ef4444 | 239, 68, 68 |
| Mode | Hex | RGB |
|---|---|---|
| Gap-Focused | #3b82f6 | 59, 130, 246 |
| Verification | #f59e0b | 245, 158, 11 |
| Deep-Independent | #ef4444 | 239, 68, 68 |
Scores (0-100) map to a continuous color gradient:
0% ─── #ef4444 (red) ─── 50% ─── #f59e0b (amber) ─── 100% ─── #22c55e (green)
Implementation (use in both contexts):
function scoreToColor(score) {
score = Math.max(0, Math.min(100, score));
if (score < 50) {
const ratio = score / 50;
const r = 239;
const g = Math.round(68 + (158 - 68) * ratio);
const b = Math.round(68 + (11 - 68) * ratio);
return `rgb(${r}, ${g}, ${b})`;
} else {
const ratio = (score - 50) / 50;
const r = Math.round(245 - (245 - 34) * ratio);
const g = Math.round(158 + (197 - 158) * ratio);
const b = Math.round(11 + (94 - 11) * ratio);
return `rgb(${r}, ${g}, ${b})`;
}
}
When rendering multi-series charts (radar, bar, line), always use this palette in order:
| Series | Hex | Usage |
|---|---|---|
| Series 1 (Readiness) | #3b82f6 (blue-500) | Primary data track |
| Series 2 (Fit-to-Purpose) | #a855f7 (purple-500) | Secondary data track |
| Series 3 | #f59e0b (amber-500) | Tertiary |
| Series 4 | #22c55e (green-500) | Quaternary |
| Series 5 | #ef4444 (red-500) | Fifth series |
| Series 6 | #06b6d4 (cyan-500) | Sixth series |
Fill areas: 10% opacity of the series color.
| Token | Hex | Usage |
|---|---|---|
| Accent | #6366f1 (indigo-500) | Interactive elements, selection highlights, primary buttons |
| Accent hover | #4f46e5 (indigo-600) | Hover state |
Dark theme (CP artifacts in Cowork):
| Token | Hex | Tailwind |
|---|---|---|
| Background | #0f172a | slate-900 |
| Surface | #1e293b | slate-800 |
| Elevated | #334155 | slate-700 |
| Border | #475569 | slate-600 |
| Text primary | #f8fafc | slate-50 |
| Text secondary | #94a3b8 | slate-400 |
| Text muted | #64748b | slate-500 |
Light theme (HTML/PDF reports):
| Token | Hex | CSS |
|---|---|---|
| Background | #ffffff | white |
| Surface | #f9fafb | gray-50 |
| Elevated | #f3f4f6 | gray-100 |
| Border | #e5e7eb | gray-200 |
| Text primary | #111827 | gray-900 |
| Text secondary | #6b7280 | gray-500 |
| Text muted | #9ca3af | gray-400 |
Primary font: Inter — loaded via Google Fonts CDN in both contexts.
CDN: https://fonts.googleapis.com/css2?family=Inter:wght@400;500;600;700&display=swap
Monospace font: JetBrains Mono, SF Mono, Monaco, Consolas, monospace — for scores, IDs, JSON, code.
| Role | Size | Weight | Line Height | Letter Spacing |
|---|---|---|---|---|
| Page title (h1) | 24px / 1.5rem | 700 | 1.2 | -0.025em |
| Section heading (h2) | 18px / 1.125rem | 600 | 1.3 | -0.02em |
| Subsection (h3) | 15px / 0.9375rem | 600 | 1.4 | -0.01em |
| Body | 14px / 0.875rem | 400 | 1.5 | 0 |
| Small / caption | 12px / 0.75rem | 500 | 1.4 | 0 |
| Micro (badges, labels) | 11px / 0.6875rem | 600 | 1.2 | 0.05em |
| Score display (large) | 32px / 2rem | 700 | 1.1 | -0.03em |
| Score display (inline) | 14px / 0.875rem | 600 | 1.2 | 0 (monospace) |
| Token | Value | Usage |
|---|---|---|
| xs | 4px | Badge padding, inline gaps |
| sm | 8px | Tight element gaps |
| md | 12px | Standard card padding, section gaps |
| lg | 16px | Section padding |
| xl | 24px | Section separation |
| 2xl | 32px | Major section breaks |
| 3xl | 48px | Page-level section separation |
| Token | Value | Usage |
|---|---|---|
| sm | 4px | Badges, pills, small chips |
| md | 6px | Buttons, inputs |
| lg | 8px | Cards, panels |
| xl | 12px | Hero badges, feature cards |
| full | 9999px | Circular elements, pills |
These proportions are consistent across both themes:
| Component | Min Height | Padding | Border Width |
|---|---|---|---|
| Determination badge (hero) | 80px | 24px 32px | 2px |
| Determination badge (inline) | 28px | 4px 12px | 1px |
| Score bar | 8px track | — | 0 (rounded-full) |
| Score ring (SVG) | 120px diameter | — | 6px stroke |
| Card (standard) | 60px | 16px | 1px |
| Card (feature/selection) | 120px | 20px 24px | 2px |
| Tab bar | 48px | 12px 20px | 3px bottom (active) |
| Table header row | 44px | 12px 16px | 2px bottom |
| Table data row | 40px | 12px 16px | 1px bottom |
| Button (primary) | 36px | 8px 16px | 0 |
| Button (secondary) | 36px | 8px 16px | 1px |
| Input / textarea | 36px | 8px 12px | 1px |
| Badge / pill (severity) | 22px | 2px 8px | 0 |
These define the visual appearance of shared components. Implementation differs (React/Tailwind vs raw HTML/CSS) but the rendered result must be visually identical.
┌──────────────────────────────────────┐
│ [ICON/CIRCLE] │
│ │
│ CONDITIONAL GO │ ← Score display size, bold
│ │
│ Company Name · 2026-03-15 │ ← Small, secondary text
└──────────────────────────────────────┘
[Label] ████████████████░░░░░░░ 78%
[CRITICAL] [SIGNIFICANT] [MODERATE] [MINOR]
All charts (Chart.js or Recharts) must use:
rgba(15, 23, 42, 0.9)), white text, 12px padding, rounded cornersresponsive: true, maintainAspectRatio: true| Data Pattern | Chart Type | When to Use |
|---|---|---|
| Multi-domain comparison (2 tracks) | Radar | Domain scores overview |
| Single-dimension comparison | Horizontal bar | Gap severity distribution, domain ranking |
| Side-by-side comparison | Grouped vertical bar | Readiness vs Fit per domain |
| Part-of-whole | Donut/Pie | Mode distribution, determination probability |
| Trend / timeline | Line | Score progression, scenario curves |
| Dense multi-variable overview | Heatmap grid (CSS) | 10-domain at-a-glance |
| Probability distribution | Histogram | Monte Carlo results |
Agents choose which charts best serve each case. These are guidelines, not mandates — if a case calls for a visualization not listed here, create it using the design system colors and typography.
This is the minimum quality bar for all outputs. Agents have full creative freedom on content structure and emphasis — but every output must meet these standards.
These are the benchmarks. Outputs should be indistinguishable in quality from reports produced by:
This means:
Information architecture:
Narrative quality:
Data visualization:
Formatting:
The same data is framed differently based on the assessor's world:
| Assessor Type | Tone | Emphasis | Language |
|---|---|---|---|
| Venture Capital | Forward-looking, opportunity-focused | Market size, team, growth potential, defensibility | "Addressable market", "burn multiple", "product-market fit" |
| Angel Investor | Founder-focused, conviction-driven | People, vision, early traction, capital efficiency | "Founder-market fit", "conviction", "inflection point" |
| Private Equity | Performance-focused, value-creation | EBITDA, margins, operational levers, de-risking | "Value creation plan", "multiple expansion", "operating leverage" |
| Credit / Debt | Risk-focused, downside-protective | Cash flow, debt service, collateral, covenants | "Coverage ratios", "debt service capacity", "stress scenario" |
| Corporate Strategic | Synergy-focused, integration-aware | Strategic fit, technology stack, talent, market access | "Synergy realization", "integration risk", "strategic value" |
| Family Office | Relationship-focused, values-aligned | Capital preservation, alignment, long-term value | "Generational wealth", "alignment with values", "capital preservation" |
| Sovereign Wealth | Mandate-focused, policy-aligned | National priorities, technology transfer, scale | "Strategic mandate", "national interest", "knowledge transfer" |
| Accelerator | Potential-focused, coaching-oriented | Coachability, founder quality, market timing | "Founder potential", "velocity of learning", "coaching leverage" |
The following limitations are inherent to the current system and are documented transparently rather than remediated. These must be disclosed in every report's Limitations section.
| Limitation | Standard | Status | Mitigation |
|---|---|---|---|
| No outcome validation — The scoring methodology has not been validated against actual investment outcomes (back-tested). | IOSCO CRA Code (annual methodology validation) | Acknowledged | Methodology is systematically constructed and peer-reviewed against professional standards. Outcome validation requires post-investment tracking data not currently available. |
| Audit trail is session-based — The audit trail is maintained in the Cowork session state and exported to the final PDF. It is not stored in an append-only immutable log during the session. | SOC 2 Type II (processing integrity), ISAE 3402 | Acknowledged | Audit trail is sealed at CP5 and embedded in the final PDF appendix. Within-session integrity depends on the Cowork platform's session management. Post-session, the PDF audit trail is the authoritative record. |
| No real-time monitoring — Assessments are point-in-time snapshots. No continuous monitoring of the assessed company occurs post-assessment. | COSO ERM (monitoring and review) | Out of scope | The tool's scope ends at recommendation delivery. Post-investment monitoring is the assessor's responsibility. Re-assessment is available via /pre-assess at any time. |
| AI model limitations — The AI model has a training knowledge cutoff and may not reflect the most recent market developments, regulatory changes, or industry events. | EU AI Act Article 15 (accuracy, robustness) | Acknowledged | All scoring-eligible evidence is obtained via live retrieval (not training knowledge). The 3H Principle and Training-Derived exclusion rule mitigate this risk. |
| Single-assessor workflow — The current workflow supports a single human assessor. Multi-party IC committee review, formal red-team panels, and multi-assessor consensus are not natively supported. | PE IC conventions (committee review) | Partial | The red-team challenge (Step 3b) provides an adversarial perspective. The final PDF is designed for IC circulation and can be reviewed by a committee externally. |
interactive-review (CP artifacts)#22c55e → green-500, #3b82f6 → blue-500, etc.html-dashboard (final reports)<style> block:root {
/* Determination */
--color-go: #22c55e;
--color-conditional-go: #3b82f6;
--color-conditional-hold: #f59e0b;
--color-no-go: #ef4444;
/* Severity */
--color-critical: #ef4444;
--color-significant: #f97316;
--color-moderate: #f59e0b;
--color-minor: #94a3b8;
/* Confidence */
--color-confidence-high: #22c55e;
--color-confidence-medium: #f59e0b;
--color-confidence-low: #ef4444;
/* Assessment modes */
--color-gap-focused: #3b82f6;
--color-verification: #f59e0b;
--color-deep-independent: #ef4444;
/* Chart series */
--color-series-1: #3b82f6;
--color-series-2: #a855f7;
--color-series-3: #f59e0b;
--color-series-4: #22c55e;
--color-series-5: #ef4444;
--color-series-6: #06b6d4;
/* Accent */
--color-accent: #6366f1;
--color-accent-hover: #4f46e5;
/* Surfaces (light theme) */
--surface-bg: #ffffff;
--surface-card: #f9fafb;
--surface-elevated: #f3f4f6;
--surface-border: #e5e7eb;
--text-primary: #111827;
--text-secondary: #6b7280;
--text-muted: #9ca3af;
/* Typography */
--font-sans: 'Inter', system-ui, -apple-system, sans-serif;
--font-mono: 'JetBrains Mono', 'SF Mono', Monaco, Consolas, monospace;
}
This design system is implemented by:
skills/interactive-review/SKILL.md — CP artifact rendering (React/Tailwind)skills/html-dashboard/SKILL.md — Final report rendering (HTML/CSS/Chart.js)skills/html-dashboard/references/chart-patterns.md — Chart implementationsskills/html-dashboard/references/component-library.md — HTML/CSS component librarynpx claudepluginhub yazo1968/market --plugin startup-assessmentBenchmarks design systems against industry standards and public systems like Material Design, Polaris, and Carbon, producing qualitative comparisons across maturity dimensions with named references.
Audits design systems for component coverage, token consistency, documentation quality, accessibility, and adoption health. Produces structured report with health scores and remediation roadmap.
Senior-level UI/UX design expert for building data-driven, premium production interfaces. Use when you need to: 1. Design complex applications (dashboards, SaaS, AI tools) from scratch 2. Generate comprehensive design systems (tokens, palettes, typography) 3. Audit existing UI for quality, accessibility, and "craft" 4. Search for proven real-world design patterns and implementation details Trigger: "design a...", "audit this...", "create a design system", "find icons", "fintech dashboard", "landing page"