Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Open Design from the CLI: driving the full design workflow with the `od` command — scripted, composable, agent-ready. Built as a decision-grade AI literacy deck for developers, power users.
.claude/skills/nexu-io-ve-terminal-mono/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-01 | ✗→✓ | ▲ Improved | 28% | 0% |
| case-02 | ✗→✓ | ▲ Improved | 28% | 0% |
| case-04 | ✗→✓ | ▲ Improved | 29% | 0% |
| case-05 | ✗→✓ | ▲ Improved | 29% | 0% |
| case-06 | ✗→✓ | ▲ Improved | 29% | 0% |
A locked, single-theme deck plugin derived from the MIT-licensed nicobailon/visual-explainer slide-deck template, restyled with the Terminal Mono aesthetic from its references/slide-patterns.md. Developer-native and typographically strict: near-black ground, monospace everything, one green voice, a faint dot grid.
Zero skeuomorphism, zero effects. No $ prompts, no red/yellow/green window chrome, no CRT scanlines, no vignettes, no glows. This theme stands on typographic discipline and the dot-grid texture alone. (The skeuomorphic "terminal theater" look belongs to a different plugin — do not drift there.)
Start from example.html in this plugin folder. It is the proven seed: copy its :root tokens (both themes), deck/slide CSS, dot-grid treatment, and the entire SlideEngine script verbatim, then replace only the slide content. Do not rewrite the design system, and do not introduce any color or font outside this spec.
:root)Dark is the default theme; light mode switches automatically via prefers-color-scheme: light. Both blocks are part of the locked spec — dual-theme support is the visual-explainer family signature and must ship in every deck.
css:root { --font-body: 'Geist Mono', 'SF Mono', Consolas, monospace; --font-mono: 'Geist Mono', 'SF Mono', Consolas, monospace; --bg: #0a0e14; /* near-black ground */ --surface: #12161e; --surface2: #1a1f2a; --surface-elevated: #222836; --border: rgba(80, 250, 123, 0.06); /* low-opacity green hairline */ --border-bright: rgba(80, 250, 123, 0.12); --text: #c8d6e5; --text-dim: #7d8c9b; /* AA on --bg for body text */ --accent: #50fa7b; /* Dracula green — the only voice */ --accent-dim: rgba(80, 250, 123, 0.08); --code-bg: #060a10; --code-text: #c8d6e5; --dot: rgba(80, 250, 123, 0.05); /* dot-grid ink */ } @media (prefers-color-scheme: light) { :root { --bg: #f4f6f8; --surface: #ffffff; --surface2: #eaecf0; --surface-elevated: #f8f9fa; --border: rgba(0, 80, 40, 0.08); --border-bright: rgba(0, 80, 40, 0.16); --text: #1a2332; --text-dim: #5a6a7a; --accent: #0d7a3e; /* deep green for light ground */ --accent-dim: rgba(13, 122, 62, 0.08); --code-bg: #1a2332; /* code blocks stay dark in light mode */ --code-text: #c8d6e5; --dot: rgba(13, 122, 62, 0.07); } }
Fonts (https://fonts.googleapis.com/css2?family=Geist+Mono:wght@400;500;600;700&display=swap). Fallback stack 'SF Mono', Consolas, monospace. No second typeface, ever.
clamp(36px, 7vw, 88px) weight 400 letter-spacing -2px; headings clamp(24px, 4vw, 42px) weight 400 letter-spacing -1px.
accent color. Subtitles: uppercase mono in --text-dim.
--text-dim.font-variant-numeric: tabular-nums.background-image: radial-gradient(var(--dot) 1px, transparent 1px); background-size: 24px 24px; This replaces the radial accent glows other family themes use — no glows here.
--border / --border-bright.No box-shadows, no blurs (except the nav-dots backdrop blur from the seed).
squares in --accent (no round dots); deck dots are squares.
lines at ~0.25 opacity forming an L in a corner, as inline SVG.
accent color at 0.05 opacity, centered behind the heading.
.deck is a 100dvh scroll-snap container; each page is one<section class="slide"> (also 100dvh, scroll-snap-align: start).
progress bar, square dot rail (right, backdrop-blurred), n / N counter, fading hints. Keyboard ←/→, ↑/↓, Space, PageUp/PageDown, Home/End; touch swipe (50px threshold); IntersectionObserver toggles .visible for entrance reveals.
#/<n> viahistory.replaceState; deep links restore on load and hashchange.
.reveal children stagger in at 0.1s steps when the slide gains.visible; animate only opacity/transform; prefers-reduced-motion disables all of it.
title, uppercase sub-line.
restrained line-art SVG aside right (1px strokes, accent, low opacity).
--surface /--surface2), 1px --border-bright divider; "before" label in --text-dim, "after" label in --accent.
.pipe-node boxes (1px hairline,mono name + uppercase meta) joined by 1px connector lines with CSS-triangle arrowheads and tiny uppercase edge labels; accent-bordered nodes mark emphasis; legend row below. Never mermaid, never any external JS.
uppercase label, dim trend line.
--surface2,zebra rows, accent-tint hover, inline code chips in --accent-dim.
--code-bg block with accent filename tab; .hl accent and.cm dim spans for highlighting (code stays dark in both themes).
> glyph in accent at 0.08 opacity asthe quote mark (no curly serif quotes), upright mono blockquote, uppercase cite.
<em> word,top hairline, corner bracket, uppercase next-steps sub-line.
Fonts stylesheet may be external. No Chart.js, no mermaid, no CDN scripts, no remote images; icons and diagram art are inline SVG or pure CSS.
when the user picks one. Split content into more slides instead of shrinking type or letting a slide scroll.
#50fa7b dark / #0d7a3e light),the listed surfaces and text tones. No red/blue/yellow semantic colors — express "before/risk/off" with --text-dim and "after/win/on" with --accent.
emphasis, never for titles.
exists.
example.html; replace content only — never redesign the theme.| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-01 | fail→pass | 50,714 | 52,705 | +4% | 1 | 1 | 0% | 8,273 | 10,622 | +28% | 0 | 0 | — |
case-02 | fail→pass | 38,137 | 35,811 | -6% | 1 | 1 | 0% | 8,255 | 10,604 | +28% | 0 | 0 | — |
case-03 | fail→fail | 40,351 | 53,037 | +31% | 1 | 1 | 0% | 8,253 | 10,602 | +28% | 0 | 0 | — |
case-04 | fail→pass | 68,745 | 50,312 | -27% | 1 | 1 | 0% | 8,232 | 10,581 | +29% | 0 | 0 | — |
case-05 | fail→pass | 44,443 | 35,868 | -19% | 1 | 1 | 0% | 8,226 | 10,575 | +29% | 0 | 0 | — |
case-06 | fail→pass | 60,527 | 48,738 | -19% | 1 | 1 | 0% | 8,229 | 10,578 | +29% | 0 | 0 | — |
case-07 | fail→pass | 51,059 | 35,491 | -30% | 1 | 1 | 0% | 8,223 | 10,572 | +29% | 0 | 0 | — |
case-08 | fail→pass | 34,784 | 48,153 | +38% | 1 | 1 | 0% | 7,870 | 10,589 | +35% | 0 | 0 | — |
case-09 | fail→fail | 39,272 | 49,378 | +26% | 1 | 1 | 0% | 8,215 | 10,564 | +29% | 0 | 0 | — |
case-10 | fail→pass | 31,884 | 36,460 | +14% | 1 | 1 | 0% | 7,055 | 10,581 | +50% | 0 | 0 | — |
case-11 | fail→pass | 28,603 | 36,888 | +29% | 1 | 1 | 0% | 6,157 | 10,567 | +72% | 0 | 0 | — |
case-12 | fail→fail | 49,321 | 38,710 | -22% | 1 | 1 | 0% | 8,219 | 10,568 | +29% | 0 | 0 | — |
case-13 | fail→pass | 39,714 | 68,555 | +73% | 1 | 1 | 0% | 8,221 | 10,570 | +29% | 0 | 0 | — |
case-14 | fail→pass | 50,778 | 34,949 | -31% | 1 | 1 | 0% | 8,223 | 10,572 | +29% | 0 | 0 | — |
case-15 | fail→fail | 49,161 | 39,481 | -20% | 1 | 1 | 0% | 8,225 | 10,574 | +29% | 0 | 0 | — |
case-16 | fail→fail | 29,234 | 36,695 | +26% | 1 | 1 | 0% | 5,937 | 10,572 | +78% | 0 | 0 | — |
case-17 | fail→fail | 37,814 | 52,743 | +39% | 1 | 1 | 0% | 8,224 | 10,573 | +29% | 0 | 0 | — |
case-18 | fail→fail | 50,037 | 38,022 | -24% | 1 | 1 | 0% | 8,219 | 10,568 | +29% | 0 | 0 | — |
case-19 | fail→pass | 24,079 | 26,545 | +10% | 1 | 1 | 0% | 5,382 | 8,363 | +55% | 0 | 0 | — |
case-20 | pass→fail | 14,744 | 38,284 | +160% | 1 | 1 | 0% | 2,232 | 10,572 | +374% | 0 | 0 | — |
case-21 | pass→fail | 17,005 | 35,209 | +107% | 1 | 1 | 0% | 3,688 | 10,562 | +186% | 0 | 0 | — |
case-22 | pass→fail | 16,896 | 38,065 | +125% | 1 | 1 | 0% | 2,803 | 10,559 | +277% | 0 | 0 | — |
DecimalAI ran this skill against gemini-3.6-flash twice over the same eval suite — once with the skill loaded and once without — and compared the two runs case by case. 22 cases were attempted. The headline lift of +41 percentage points is the difference between those two pass rates over the 22 comparable cases. 3 cases got worse with the skill loaded, and they are included in that figure.
Without the skill loaded, the model failed this case. With it loaded, the same prompt on the same model passed. This is one improved case from the latest verified run; every case, including any that regressed, is in the table above.
Other measured skills in the registry, with their headline benchmark lift.