Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Generate documentation following the Diataxis framework
.claude/skills/a5c-ai-diataxis-doc-generator/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-06 | ✗→✓ | ▲ Improved | 2% | 0% |
| case-11 | ✗→✓ | ▲ Improved | 16% | 0% |
| case-01 | ✓→✓ | = Same ✓ | 24% | 0% |
| case-02 | ✓→✓ | = Same ✓ | 109% | 0% |
| case-04 | ✓→✓ | = Same ✓ | 30% | 0% |
This skill generates comprehensive documentation following the Diataxis framework, organizing content into tutorials, how-to guides, reference, and explanation categories for optimal developer learning.
yamlskill: name: diataxis-doc-generator context: apiSpec: ./openapi.yaml platform: docusaurus languages: - javascript - python - curl includePlayground: true versioning: true
| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-01 | pass→pass | 9,813 | 9,003 | -8% | 1 | 1 | 0% | 844 | 1,050 | +24% | 0 | 0 | — |
case-02 | pass→pass | 3,875 | 11,354 | +193% | 1 | 1 | 0% | 773 | 1,616 | +109% | 0 | 0 | — |
case-03 | fail→fail | 5,487 | 4,436 | -19% | 1 | 1 | 0% | 1,032 | 1,175 | +14% | 0 | 0 | — |
case-04 | pass→pass | 19,500 | 19,198 | -2% | 1 | 1 | 0% | 2,233 | 2,895 | +30% | 0 | 0 | — |
case-05 | pass→pass | 15,606 | 13,807 | -12% | 1 | 1 | 0% | 1,682 | 1,883 | +12% | 0 | 0 | — |
case-06 | fail→pass | 12,812 | 16,013 | +25% | 1 | 1 | 0% | 2,206 | 2,260 | +2% | 0 | 0 | — |
case-07 | pass→pass | 13,827 | 14,850 | +7% | 1 | 1 | 0% | 1,334 | 2,066 | +55% | 0 | 0 | — |
case-08 | pass→pass | 10,442 | 11,366 | +9% | 1 | 1 | 0% | 879 | 1,326 | +51% | 0 | 0 | — |
case-09 | pass→pass | 11,228 | 16,470 | +47% | 1 | 1 | 0% | 1,880 | 2,317 | +23% | 0 | 0 | — |
case-10 | pass→pass | 10,682 | 11,415 | +7% | 1 | 1 | 0% | 1,005 | 1,370 | +36% | 0 | 0 | — |
case-11 | fail→pass | 13,126 | 19,887 | +52% | 1 | 1 | 0% | 2,163 | 2,507 | +16% | 0 | 0 | — |
case-12 | pass→pass | 4,976 | 10,758 | +116% | 1 | 1 | 0% | 793 | 1,184 | +49% | 0 | 0 | — |
case-13 | pass→pass | 11,992 | 12,391 | +3% | 1 | 1 | 0% | 1,962 | 2,549 | +30% | 0 | 0 | — |
case-14 | pass→pass | 12,862 | 12,574 | -2% | 1 | 1 | 0% | 1,260 | 1,692 | +34% | 0 | 0 | — |
case-15 | pass→pass | 11,292 | 11,220 | -1% | 1 | 1 | 0% | 1,082 | 1,493 | +38% | 0 | 0 | — |
case-16 | pass→pass | 15,386 | 12,751 | -17% | 1 | 1 | 0% | 1,630 | 2,470 | +52% | 0 | 0 | — |
case-17 | pass→pass | 15,843 | 24,572 | +55% | 1 | 1 | 0% | 2,648 | 3,632 | +37% | 0 | 0 | — |
case-18 | pass→pass | 16,126 | 16,387 | +2% | 1 | 1 | 0% | 1,811 | 2,249 | +24% | 0 | 0 | — |
case-19 | pass→pass | 20,641 | 17,034 | -17% | 1 | 1 | 0% | 2,521 | 2,374 | -6% | 0 | 0 | — |
case-20 | pass→pass | 16,755 | 10,069 | -40% | 1 | 1 | 0% | 1,912 | 2,108 | +10% | 0 | 0 | — |
case-21 | pass→pass | 17,270 | 17,547 | +2% | 1 | 1 | 0% | 1,891 | 2,307 | +22% | 0 | 0 | — |
case-22 | pass→pass | 16,637 | 17,553 | +6% | 1 | 1 | 0% | 1,936 | 2,215 | +14% | 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 +9 percentage points is the difference between those two pass rates over the 22 comparable cases.
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.