Loading skill
Install any skill in seconds. Free to start, no credit card required.
Get Started Free →A workplace-safety compliance review for a manufacturing regulator — findings, the evidence chain, and the corrective mandate. Built as a decision-grade policy briefing deck for regulator, plant leadership.
.claude/skills/nexu-io-html-ppt-zhangzara-stencil-tablet/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-22 | ✗→✓ | ▲ Improved | 9% | 0% |
| case-02 | ✗→✓ | ▲ Improved | 9% | 0% |
| case-15 | ✗→✓ | ▲ Improved | 9% | 0% |
| case-20 | ✗→✓ | ▲ Improved | 9% | 0% |
| case-06 | ✓→✗ | ▼ Worse | 229% | 0% |
> Bone paper with stencil-cut headlines and a six-color earth palette: archaeology meets brand.
A single self-contained HTML deck — typography, palette, decorative system, and slide vocabulary are all tuned together. Mixing layouts across templates breaks the system; stay inside this one.
Anything that should feel archival, tactile, and weighty-graphic: museum and cultural-institution decks, art / architecture brands, longform research, heritage and craft brands, manifestos. A great choice anytime — including across tech and business — when you want the deck to feel like a field manual rather than a slide deck.
Contexts that demand digital-native polish or playful pop — the stencil-cut display and earth-tone palette commit to a deliberate analog feel.
example.html AND the assets/ folder into the user's workspace.This template ships an assets/deck-stage.js runtime (keyboard navigation, stage rendering); the HTML references it as assets/deck-stage.js, so the file must sit next to the cloned HTML or that path will 404 in the generated artifact and navigation will silently break. Inlining the JS into a single <script> block in the HTML is an acceptable alternative when a single self-contained file is preferred.
numbers, names, dates, and section labels. Match existing dimensions when swapping image placeholders.
restructure the layout grid, or strip decorative elements (corner brackets, paper grain, geometric shapes, illustrated SVGs). They are part of the identity.
than the demo holds, duplicate an existing slide of the most appropriate layout. If less, drop slides from the bottom. Update page-number labels.
doesn't have, design it from scratch using the same fonts, palette, decorative vocabulary, spacing rhythm, and component grammar — never bail to a different template.
assets/deck-stage.js or inline keyboard handler, leave it intact.
Emit between <artifact> tags:
<artifact identifier="zhangzara-stencil-tablet" type="text/html" title="Deck Title">
<!doctype html>
<html>...</html>
</artifact>Vendored from upstream MIT-licensed zarazhangrui/beautiful-html-templates.
The full upstream MIT license text — including the original copyright notice — ships in this skill at LICENSE and must be redistributed alongside any copy of example.html, template.json, or any vendored assets/ runtime. See template.json for the upstream metadata snapshot.
| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-22 | fail→pass | 36,723 | 43,544 | +19% | 1 | 1 | 0% | 8,222 | 8,972 | +9% | 0 | 0 | — |
case-01 | fail→fail | 101,432 | 41,192 | -59% | 1 | 1 | 0% | 8,259 | 9,009 | +9% | 0 | 0 | — |
case-02 | fail→pass | 43,364 | 53,108 | +22% | 1 | 1 | 0% | 8,247 | 8,997 | +9% | 0 | 0 | — |
case-03 | fail→fail | 53,115 | 42,929 | -19% | 1 | 1 | 0% | 8,240 | 8,990 | +9% | 0 | 0 | — |
case-04 | pass→pass | 49,738 | 40,567 | -18% | 1 | 1 | 0% | 8,235 | 8,985 | +9% | 0 | 0 | — |
case-05 | pass→pass | 18,850 | 24,064 | +28% | 1 | 1 | 0% | 2,790 | 4,332 | +55% | 0 | 0 | — |
case-06 | pass→fail | 17,054 | 41,914 | +146% | 1 | 1 | 0% | 2,726 | 8,982 | +229% | 0 | 0 | — |
case-07 | fail→fail | 51,688 | 41,785 | -19% | 1 | 1 | 0% | 8,236 | 8,986 | +9% | 0 | 0 | — |
case-08 | fail→fail | 37,450 | 67,378 | +80% | 1 | 1 | 0% | 8,228 | 8,978 | +9% | 0 | 0 | — |
case-09 | fail→fail | 43,571 | 91,805 | +111% | 1 | 1 | 0% | 8,232 | 8,982 | +9% | 0 | 0 | — |
case-10 | fail→fail | 36,870 | 64,109 | +74% | 1 | 1 | 0% | 8,217 | 8,967 | +9% | 0 | 0 | — |
case-11 | fail→fail | 41,336 | 49,710 | +20% | 1 | 1 | 0% | 8,234 | 8,984 | +9% | 0 | 0 | — |
case-12 | fail→fail | 40,238 | 43,302 | +8% | 1 | 1 | 0% | 8,229 | 8,979 | +9% | 0 | 0 | — |
case-13 | fail→fail | 32,572 | 51,957 | +60% | 1 | 1 | 0% | 6,560 | 8,971 | +37% | 0 | 0 | — |
case-14 | fail→fail | 72,741 | 52,682 | -28% | 1 | 1 | 0% | 8,218 | 8,968 | +9% | 0 | 0 | — |
case-15 | fail→pass | 51,472 | 39,834 | -23% | 1 | 1 | 0% | 8,220 | 8,970 | +9% | 0 | 0 | — |
case-16 | fail→fail | 39,642 | 41,532 | +5% | 1 | 1 | 0% | 7,748 | 8,964 | +16% | 0 | 0 | — |
case-17 | fail→fail | 40,750 | 54,153 | +33% | 1 | 1 | 0% | 8,229 | 8,979 | +9% | 0 | 0 | — |
case-18 | fail→fail | 54,650 | 41,247 | -25% | 1 | 1 | 0% | 8,220 | 8,970 | +9% | 0 | 0 | — |
case-19 | pass→pass | 63,679 | 39,310 | -38% | 1 | 1 | 0% | 8,233 | 8,983 | +9% | 0 | 0 | — |
case-20 | fail→pass | 76,722 | 53,804 | -30% | 1 | 1 | 0% | 8,227 | 8,977 | +9% | 0 | 0 | — |
case-21 | fail→fail | 42,248 | 43,943 | +4% | 1 | 1 | 0% | 8,220 | 8,970 | +9% | 0 | 0 | — |
case-23 | fail→fail | 41,751 | 40,597 | -3% | 1 | 1 | 0% | 8,216 | 8,966 | +9% | 0 | 0 | — |
case-24 | fail→fail | 40,921 | 40,924 | +0% | 1 | 1 | 0% | 8,224 | 8,974 | +9% | 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. 24 cases were attempted. The headline lift of +13 percentage points is the difference between those two pass rates over the 24 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.