Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Creates interactive HTML playgrounds — self-contained single-file explorers that let users configure something visually through controls, see a live preview, and copy out a prompt. Use when the user asks to make a playground, explorer, or interactive tool for a topic.
.claude/skills/anthropics-playground/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-04 | ✓→✗ | ▼ Worse | 30% | 0% |
| case-05 | ✓→✗ | ▼ Worse | 489% | 0% |
| case-10 | ✓→✗ | ▼ Worse | 13% | 0% |
| case-12 | ✓→✗ | ▼ Worse | 29% | 0% |
| case-15 | ✓→✗ | ▼ Worse | 13% | 0% |
A playground is a self-contained HTML file with interactive controls on one side, a live preview on the other, and a prompt output at the bottom with a copy button. The user adjusts controls, explores visually, then copies the generated prompt back into Claude.
When the user asks for an interactive playground, explorer, or visual tool for a topic — especially when the input space is large, visual, or structural and hard to express as plain text.
templates/:templates/design-playground.md — Visual design decisions (components, layouts, spacing, color, typography)templates/data-explorer.md — Data and query building (SQL, APIs, pipelines, regex)templates/concept-map.md — Learning and exploration (concept maps, knowledge gaps, scope mapping)templates/document-critique.md — Document review (suggestions with approve/reject/comment workflow)templates/diff-review.md — Code review (git diffs, commits, PRs with line-by-line commenting)templates/code-map.md — Codebase architecture (component relationships, data flow, layer diagrams)open <filename>.html to launch it in the user's default browser.Keep a single state object. Every control writes to it, every render reads from it.
javascriptconst state = { /* all configurable values */ }; function updateAll() { renderPreview(); // update the visual updatePrompt(); // rebuild the prompt text } // Every control calls updateAll() on change
javascriptfunction updatePrompt() { const parts = []; // Only mention non-default values if (state.borderRadius !== DEFAULTS.borderRadius) { parts.push(`border-radius of ${state.borderRadius}px`); } // Use qualitative language alongside numbers if (state.shadowBlur > 16) parts.push('a pronounced shadow'); else if (state.shadowBlur > 0) parts.push('a subtle shadow'); prompt.textContent = `Update the card to use ${parts.join(', ')}.`; }
| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-01 | fail→fail | 25,098 | 10,467 | -58% | 1 | 1 | 0% | 6,229 | 2,950 | -53% | 0 | 0 | — |
case-02 | fail→fail | 27,026 | 23,513 | -13% | 1 | 1 | 0% | 6,212 | 7,040 | +13% | 0 | 0 | — |
case-03 | fail→fail | 26,970 | 26,959 | -0% | 1 | 1 | 0% | 6,210 | 7,038 | +13% | 0 | 0 | — |
case-04 | pass→fail | 23,433 | 24,706 | +5% | 1 | 1 | 0% | 5,384 | 6,998 | +30% | 0 | 0 | — |
case-05 | pass→fail | 5,927 | 24,715 | +317% | 1 | 1 | 0% | 1,187 | 6,995 | +489% | 0 | 0 | — |
case-06 | pass→pass | 10,214 | 9,549 | -7% | 1 | 1 | 0% | 2,369 | 3,026 | +28% | 0 | 0 | — |
case-07 | fail→fail | 26,970 | 23,198 | -14% | 1 | 1 | 0% | 6,170 | 6,998 | +13% | 0 | 0 | — |
case-08 | pass→pass | 23,500 | 23,448 | -0% | 1 | 1 | 0% | 6,172 | 7,000 | +13% | 0 | 0 | — |
case-09 | fail→fail | 23,129 | 9,330 | -60% | 1 | 1 | 0% | 6,164 | 2,685 | -56% | 0 | 0 | — |
case-10 | pass→fail | 24,712 | 23,386 | -5% | 1 | 1 | 0% | 6,168 | 6,996 | +13% | 0 | 0 | — |
case-11 | fail→fail | 25,050 | 24,877 | -1% | 1 | 1 | 0% | 6,169 | 6,997 | +13% | 0 | 0 | — |
case-12 | pass→fail | 21,136 | 24,235 | +15% | 1 | 1 | 0% | 5,426 | 6,998 | +29% | 0 | 0 | — |
case-13 | fail→fail | 24,814 | 11,175 | -55% | 1 | 1 | 0% | 6,163 | 2,946 | -52% | 0 | 0 | — |
case-14 | fail→fail | 24,781 | 24,412 | -1% | 1 | 1 | 0% | 6,166 | 6,995 | +13% | 0 | 0 | — |
case-15 | pass→fail | 25,154 | 25,787 | +3% | 1 | 1 | 0% | 6,167 | 6,995 | +13% | 0 | 0 | — |
case-16 | pass→pass | 24,733 | 24,897 | +1% | 1 | 1 | 0% | 6,166 | 6,994 | +13% | 0 | 0 | — |
case-17 | fail→fail | 25,483 | 25,943 | +2% | 1 | 1 | 0% | 6,164 | 6,992 | +13% | 0 | 0 | — |
case-18 | fail→fail | 27,804 | 26,221 | -6% | 1 | 1 | 0% | 6,163 | 6,991 | +13% | 0 | 0 | — |
case-19 | fail→fail | 26,539 | 25,843 | -3% | 1 | 1 | 0% | 6,163 | 6,991 | +13% | 0 | 0 | — |
case-20 | fail→fail | 27,096 | 23,926 | -12% | 1 | 1 | 0% | 6,164 | 6,993 | +13% | 0 | 0 | — |
case-21 | fail→fail | 27,272 | 24,923 | -9% | 1 | 1 | 0% | 6,163 | 6,991 | +13% | 0 | 0 | — |
case-22 | fail→fail | 25,175 | 25,598 | +2% | 1 | 1 | 0% | 6,162 | 6,990 | +13% | 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 -36 percentage points is the difference between those two pass rates over the 22 comparable cases. 7 cases got worse with the skill loaded, and they are included in that figure.
Other measured skills in the registry, with their headline benchmark lift.