Loading skill
Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Manages skill version history and automated improvement cycles. Use for skil lmaintenance or user says evolve skills, version skills. Use when the user n eeds this capability. Triggers: evolve skills, version skills, skill mainten ance.
.claude/skills/stijnman-skill-evolution-engine/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-01 | ✗→✓ | ▲ Improved | -29% | 0% |
| case-02 | ✗→✓ | ▲ Improved | -21% | 0% |
| case-03 | ✗→✓ | ▲ Improved | -36% | 0% |
| case-06 | ✗→✓ | ▲ Improved | -31% | 0% |
| case-08 | ✗→✓ | ▲ Improved | -46% | 0% |
skill-evolverevolutionevolverhyper-skill-tester| Failure | Response | |---------|----------| | Score regression | Preserve the evidence, present the latest backup and rollback diff, and wait for user approval before restoring. |
Input: User request matching triggers above. Output: Structured result per workflow with integrations invoked as needed.
| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-01 | fail→pass | 13,312 | 12,141 | -9% | 1 | 1 | 0% | 2,466 | 1,739 | -29% | 0 | 0 | — |
case-02 | fail→pass | 19,238 | 10,086 | -48% | 1 | 1 | 0% | 2,407 | 1,909 | -21% | 0 | 0 | — |
case-03 | fail→pass | 20,455 | 7,735 | -62% | 1 | 1 | 0% | 2,395 | 1,542 | -36% | 0 | 0 | — |
case-04 | fail→fail | 16,188 | 11,940 | -26% | 1 | 1 | 0% | 2,624 | 2,256 | -14% | 0 | 0 | — |
case-05 | fail→fail | 18,890 | 22,665 | +20% | 1 | 1 | 0% | 3,265 | 3,704 | +13% | 0 | 0 | — |
case-06 | fail→pass | 9,661 | 4,463 | -54% | 1 | 1 | 0% | 1,467 | 1,018 | -31% | 0 | 0 | — |
case-07 | fail→fail | 16,330 | 15,241 | -7% | 1 | 1 | 0% | 2,926 | 2,943 | +1% | 0 | 0 | — |
case-08 | fail→pass | 13,827 | 3,703 | -73% | 1 | 1 | 0% | 1,742 | 934 | -46% | 0 | 0 | — |
case-09 | fail→fail | 14,085 | 4,247 | -70% | 1 | 1 | 0% | 1,755 | 901 | -49% | 0 | 0 | — |
case-10 | fail→pass | 7,597 | 4,594 | -40% | 1 | 1 | 0% | 1,074 | 1,050 | -2% | 0 | 0 | — |
case-11 | pass→pass | 9,668 | 4,735 | -51% | 1 | 1 | 0% | 1,524 | 951 | -38% | 0 | 0 | — |
case-12 | pass→pass | 15,413 | 4,858 | -68% | 1 | 1 | 0% | 2,296 | 1,039 | -55% | 0 | 0 | — |
case-13 | fail→pass | 9,380 | 1,643 | -82% | 1 | 1 | 0% | 1,419 | 547 | -61% | 0 | 0 | — |
case-14 | pass→pass | 9,524 | 2,154 | -77% | 1 | 1 | 0% | 1,376 | 498 | -64% | 0 | 0 | — |
case-15 | fail→pass | 8,462 | 3,010 | -64% | 1 | 1 | 0% | 1,337 | 777 | -42% | 0 | 0 | — |
case-16 | pass→pass | 9,575 | 4,191 | -56% | 1 | 1 | 0% | 1,501 | 813 | -46% | 0 | 0 | — |
case-17 | fail→pass | 10,807 | 4,202 | -61% | 1 | 1 | 0% | 1,607 | 868 | -46% | 0 | 0 | — |
case-18 | fail→pass | 10,083 | 4,410 | -56% | 1 | 1 | 0% | 1,505 | 961 | -36% | 0 | 0 | — |
case-19 | fail→pass | 14,717 | 7,692 | -48% | 1 | 1 | 0% | 2,231 | 1,500 | -33% | 0 | 0 | — |
case-20 | fail→pass | 14,067 | 3,026 | -78% | 1 | 1 | 0% | 2,100 | 654 | -69% | 0 | 0 | — |
case-21 | fail→pass | 9,498 | 2,886 | -70% | 1 | 1 | 0% | 1,416 | 800 | -44% | 0 | 0 | — |
case-22 | fail→pass | 13,732 | 2,622 | -81% | 1 | 1 | 0% | 1,849 | 635 | -66% | 0 | 0 | — |
case-23 | fail→pass | 15,949 | 6,519 | -59% | 1 | 1 | 0% | 2,078 | 1,148 | -45% | 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. 23 cases were attempted. The headline lift of +65 percentage points is the difference between those two pass rates over the 23 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.