Loading skill
Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Use this skill when working in Bun-based projects (or Node projects running on Bun) and you want reliable runtime expectations, dependency management, and debugging workflows.
.claude/skills/amariahak-bun-patterns/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-02 | ✗→✓ | ▲ Improved | -22% | 0% |
| case-15 | ✗→✓ | ▲ Improved | -36% | 0% |
| case-18 | ✗→✓ | ▲ Improved | -20% | 0% |
| case-01 | ✓→✓ | ＝ Same ✓ | -37% | 0% |
| case-03 | ✓→✓ | ＝ Same ✓ | -30% | 0% |
Use this skill when working in Bun-based projects (or Node projects running on Bun) and you want reliable runtime expectations, dependency management, and debugging workflows.
Bun aims for Node compatibility, but differences exist:
When debugging a bug:
bun.lockbAvoid mixing npm/yarn/pnpm and bun unless the project explicitly supports it.
bun.lockb and scripts; locate entrypoints.bun install and the relevant bun run <script>.| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-01 | pass→pass | 13,878 | 6,402 | -54% | 1 | 1 | 0% | 2,216 | 1,385 | -38% | 0 | 0 | — |
case-02 | fail→pass | 14,689 | 9,026 | -39% | 1 | 1 | 0% | 2,469 | 1,919 | -22% | 0 | 0 | — |
case-03 | pass→pass | 19,896 | 6,453 | -68% | 1 | 1 | 0% | 1,915 | 1,350 | -30% | 0 | 0 | — |
case-04 | pass→pass | 13,473 | 8,377 | -38% | 1 | 1 | 0% | 2,430 | 1,771 | -27% | 0 | 0 | — |
case-05 | pass→pass | 8,206 | 2,965 | -64% | 1 | 1 | 0% | 1,239 | 730 | -41% | 0 | 0 | — |
case-06 | pass→pass | 11,506 | 4,667 | -59% | 1 | 1 | 0% | 2,046 | 1,166 | -43% | 0 | 0 | — |
case-07 | pass→pass | 5,678 | 4,531 | -20% | 1 | 1 | 0% | 1,019 | 879 | -14% | 0 | 0 | — |
case-08 | pass→pass | 13,608 | 6,229 | -54% | 1 | 1 | 0% | 2,191 | 1,276 | -42% | 0 | 0 | — |
case-09 | pass→pass | 11,257 | 3,441 | -69% | 1 | 1 | 0% | 2,026 | 924 | -54% | 0 | 0 | — |
case-10 | pass→pass | 10,852 | 5,545 | -49% | 1 | 1 | 0% | 1,780 | 852 | -52% | 0 | 0 | — |
case-11 | pass→pass | 7,033 | 3,929 | -44% | 1 | 1 | 0% | 1,303 | 1,005 | -23% | 0 | 0 | — |
case-12 | pass→pass | 14,582 | 8,741 | -40% | 1 | 1 | 0% | 2,455 | 1,697 | -31% | 0 | 0 | — |
case-13 | fail→fail | 6,462 | 2,216 | -66% | 1 | 1 | 0% | 1,140 | 659 | -42% | 0 | 0 | — |
case-14 | pass→pass | 15,564 | 7,696 | -51% | 1 | 1 | 0% | 2,423 | 1,608 | -34% | 0 | 0 | — |
case-15 | fail→pass | 11,419 | 5,390 | -53% | 1 | 1 | 0% | 1,972 | 1,254 | -36% | 0 | 0 | — |
case-16 | pass→pass | 8,175 | 3,530 | -57% | 1 | 1 | 0% | 1,313 | 821 | -37% | 0 | 0 | — |
case-17 | pass→pass | 8,899 | 4,518 | -49% | 1 | 1 | 0% | 1,696 | 1,123 | -34% | 0 | 0 | — |
case-18 | fail→pass | 9,526 | 5,936 | -38% | 1 | 1 | 0% | 1,474 | 1,180 | -20% | 0 | 0 | — |
case-19 | pass→pass | 7,600 | 2,491 | -67% | 1 | 1 | 0% | 1,242 | 716 | -42% | 0 | 0 | — |
case-20 | pass→pass | 8,990 | 9,734 | +8% | 1 | 1 | 0% | 1,666 | 1,576 | -5% | 0 | 0 | — |
case-21 | pass→pass | 8,976 | 7,894 | -12% | 1 | 1 | 0% | 1,565 | 1,637 | +5% | 0 | 0 | — |
case-22 | pass→pass | 10,795 | 7,123 | -34% | 1 | 1 | 0% | 1,961 | 1,637 | -17% | 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 +14 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.