Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Turn a topic, brainstorm, or document into a structured mind map. Use when asked to brainstorm around a theme, organize ideas, break a topic into branches, or summarize something as a mind map. Produces a ready-to-render Mermaid mindmap (renders live, exportable as PNG/SVG) plus a short note on the structure chosen.
.claude/skills/mohitagw15856-mind-map/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-19 | ✗→✓ | ▲ Improved | 21% | 0% |
| case-21 | ✓→✗ | ▼ Worse | 10% | 0% |
| case-01 | ✓→✓ | = Same ✓ | -13% | 0% |
| case-22 | ✓→✓ | = Same ✓ | 41% | 0% |
| case-03 | ✓→✓ | = Same ✓ | 12% | 0% |
A mind map turns a fuzzy topic into a branching structure you can see — central idea in the middle, themes radiating out, details hanging off each. This skill takes a topic, a brain-dump, or a document and organizes it into a clean Mermaid mindmap with sensible, balanced branches.
Ask for these only if they aren't already provided:
One line on how you structured it (the organizing principle for the main branches).
mermaidmindmap root((Central topic)) Theme A Idea A1 Idea A2 Theme B Idea B1 Idea B2 Theme C Idea C1
Structure note — why these main branches, and anything that didn't fit (parked items).
mindmap. The center is root((Text)).root(( ))).Mind-mapping practice (radial hierarchy, balanced branches, MECE-ish themes), expressed as renderable Mermaid.
| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-01 | pass→pass | 12,046 | 6,771 | -44% | 1 | 1 | 0% | 2,161 | 1,880 | -13% | 0 | 0 | — |
case-22 | pass→pass | 8,369 | 8,795 | +5% | 1 | 1 | 0% | 1,590 | 2,240 | +41% | 0 | 0 | — |
case-02 | fail→fail | 10,531 | 6,385 | -39% | 1 | 1 | 0% | 2,076 | 1,802 | -13% | 0 | 0 | — |
case-03 | pass→pass | 20,627 | 16,718 | -19% | 1 | 1 | 0% | 984 | 1,104 | +12% | 0 | 0 | — |
case-04 | pass→pass | 9,327 | 14,031 | +50% | 1 | 1 | 0% | 667 | 1,915 | +187% | 0 | 0 | — |
case-05 | pass→pass | 21,847 | 12,913 | -41% | 1 | 1 | 0% | 1,091 | 1,644 | +51% | 0 | 0 | — |
case-06 | pass→pass | 3,363 | 17,798 | +429% | 1 | 1 | 0% | 553 | 1,433 | +159% | 0 | 0 | — |
case-07 | pass→pass | 8,161 | 11,200 | +37% | 1 | 1 | 0% | 1,594 | 1,727 | +8% | 0 | 0 | — |
case-08 | pass→pass | 8,414 | 5,799 | -31% | 1 | 1 | 0% | 1,506 | 1,568 | +4% | 0 | 0 | — |
case-09 | pass→pass | 3,467 | 5,156 | +49% | 1 | 1 | 0% | 768 | 1,643 | +114% | 0 | 0 | — |
case-10 | pass→pass | 8,431 | 5,878 | -30% | 1 | 1 | 0% | 1,762 | 1,674 | -5% | 0 | 0 | — |
case-11 | pass→pass | 3,554 | 4,691 | +32% | 1 | 1 | 0% | 772 | 1,555 | +101% | 0 | 0 | — |
case-12 | fail→fail | 9,226 | 8,016 | -13% | 1 | 1 | 0% | 1,651 | 2,061 | +25% | 0 | 0 | — |
case-13 | pass→pass | 6,675 | 5,216 | -22% | 1 | 1 | 0% | 1,341 | 1,620 | +21% | 0 | 0 | — |
case-14 | fail→fail | 13,482 | 5,655 | -58% | 1 | 1 | 0% | 2,736 | 1,607 | -41% | 0 | 0 | — |
case-15 | pass→pass | 2,649 | 3,345 | +26% | 1 | 1 | 0% | 543 | 1,244 | +129% | 0 | 0 | — |
case-16 | pass→pass | 7,823 | 7,198 | -8% | 1 | 1 | 0% | 1,613 | 2,036 | +26% | 0 | 0 | — |
case-17 | fail→fail | 4,072 | 6,305 | +55% | 1 | 1 | 0% | 766 | 1,813 | +137% | 0 | 0 | — |
case-18 | fail→fail | 7,612 | 5,367 | -29% | 1 | 1 | 0% | 1,525 | 1,668 | +9% | 0 | 0 | — |
case-19 | fail→pass | 8,315 | 7,561 | -9% | 1 | 1 | 0% | 1,636 | 1,980 | +21% | 0 | 0 | — |
case-20 | pass→pass | 9,114 | 9,826 | +8% | 1 | 1 | 0% | 1,908 | 2,596 | +36% | 0 | 0 | — |
case-21 | pass→fail | 10,349 | 9,260 | -11% | 1 | 1 | 0% | 2,078 | 2,293 | +10% | 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 0 percentage points is the difference between those two pass rates over the 22 comparable cases. 1 case got worse with the skill loaded, and it is 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.