Loading skill
Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Write a polished git commit message from my staged changes following Conventional Commits, with an optional emoji and the related issue key.
.claude/skills/kainulla-my-commit-wizard/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-10 | ✓→✗ | ▼ Worse | -33% | 0% |
| case-01 | ✓→✓ | ＝ Same ✓ | -12% | 0% |
| case-02 | ✓→✓ | ＝ Same ✓ | -18% | 0% |
| case-03 | ✓→✓ | ＝ Same ✓ | -1% | 0% |
| case-04 | ✓→✓ | ＝ Same ✓ | -12% | 0% |
A sample skill used to demo skillfire. Run:
bashnode bin/skillfire.ts examples/sample-commit-skill
| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-01 | pass→pass | 3,980 | 3,135 | -21% | 1 | 1 | 0% | 668 | 591 | -12% | 0 | 0 | — |
case-02 | pass→pass | 5,159 | 2,656 | -49% | 1 | 1 | 0% | 547 | 450 | -18% | 0 | 0 | — |
case-03 | pass→pass | 3,051 | 4,110 | +35% | 1 | 1 | 0% | 490 | 484 | -1% | 0 | 0 | — |
case-04 | pass→pass | 3,268 | 2,892 | -12% | 1 | 1 | 0% | 495 | 436 | -12% | 0 | 0 | — |
case-05 | pass→pass | 4,966 | 3,163 | -36% | 1 | 1 | 0% | 608 | 543 | -11% | 0 | 0 | — |
case-06 | pass→pass | 3,836 | 2,486 | -35% | 1 | 1 | 0% | 655 | 412 | -37% | 0 | 0 | — |
case-07 | pass→pass | 4,467 | 2,585 | -42% | 1 | 1 | 0% | 727 | 459 | -37% | 0 | 0 | — |
case-13 | pass→pass | 3,427 | 2,082 | -39% | 1 | 1 | 0% | 495 | 351 | -29% | 0 | 0 | — |
case-08 | pass→pass | 3,204 | 3,397 | +6% | 1 | 1 | 0% | 605 | 633 | +5% | 0 | 0 | — |
case-09 | pass→pass | 2,829 | 2,492 | -12% | 1 | 1 | 0% | 467 | 440 | -6% | 0 | 0 | — |
case-10 | pass→fail | 3,872 | 2,358 | -39% | 1 | 1 | 0% | 618 | 416 | -33% | 0 | 0 | — |
case-11 | pass→pass | 4,282 | 3,571 | -17% | 1 | 1 | 0% | 745 | 601 | -19% | 0 | 0 | — |
case-12 | pass→pass | 9,519 | 5,793 | -39% | 1 | 1 | 0% | 1,592 | 882 | -45% | 0 | 0 | — |
case-14 | pass→pass | 2,845 | 2,763 | -3% | 1 | 1 | 0% | 404 | 484 | +20% | 0 | 0 | — |
case-15 | pass→pass | 3,254 | 2,502 | -23% | 1 | 1 | 0% | 514 | 469 | -9% | 0 | 0 | — |
case-16 | pass→pass | 3,719 | 2,580 | -31% | 1 | 1 | 0% | 666 | 498 | -25% | 0 | 0 | — |
case-17 | pass→pass | 3,853 | 4,028 | +5% | 1 | 1 | 0% | 649 | 597 | -8% | 0 | 0 | — |
case-18 | pass→pass | 4,304 | 3,845 | -11% | 1 | 1 | 0% | 781 | 680 | -13% | 0 | 0 | — |
case-19 | pass→pass | 4,064 | 3,943 | -3% | 1 | 1 | 0% | 543 | 696 | +28% | 0 | 0 | — |
case-20 | pass→pass | 4,433 | 3,385 | -24% | 1 | 1 | 0% | 811 | 659 | -19% | 0 | 0 | — |
case-21 | pass→pass | 9,436 | 4,333 | -54% | 1 | 1 | 0% | 1,356 | 811 | -40% | 0 | 0 | — |
case-22 | pass→pass | 3,716 | 2,990 | -20% | 1 | 1 | 0% | 745 | 558 | -25% | 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 -100 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.
Other measured skills in the registry, with their headline benchmark lift.