Loading skill
Install any skill in seconds. Free to start, no credit card required.
Get Started Free →BullMQ expert for Redis-backed job queues, background processing, and reliable async execution in Node.js/TypeScript applications. Use when: bullmq, bull queue, redis queue, background job, job queue.
.claude/skills/dokhacgiakhoa-bullmq-specialist/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-11 | ✗→✓ | ▲ Improved | 14% | 0% |
| case-12 | ✗→✓ | ▲ Improved | 11% | 0% |
| case-21 | ✗→✓ | ▲ Improved | 12% | 0% |
| case-01 | ✓→✗ | ▼ Worse | 1% | 0% |
| case-02 | ✓→✓ | ＝ Same ✓ | 24% | 0% |
You are a BullMQ expert who has processed billions of jobs in production. You understand that queues are the backbone of scalable applications - they decouple services, smooth traffic spikes, and enable reliable async processing.
You've debugged stuck jobs at 3am, optimized worker concurrency for maximum throughput, and designed job flows that handle complex multi-step processes. You know that most queue problems are actually Redis problems or application design problems.
Your core philosophy:
| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-01 | pass→fail | 16,590 | 27,835 | +68% | 1 | 1 | 0% | 2,327 | 2,352 | +1% | 0 | 0 | — |
case-02 | pass→pass | 10,306 | 11,442 | +11% | 1 | 1 | 0% | 1,852 | 2,288 | +24% | 0 | 0 | — |
case-03 | pass→pass | 13,123 | 11,098 | -15% | 1 | 1 | 0% | 2,433 | 2,246 | -8% | 0 | 0 | — |
case-04 | pass→pass | 12,317 | 13,224 | +7% | 1 | 1 | 0% | 2,331 | 2,852 | +22% | 0 | 0 | — |
case-05 | pass→pass | 13,831 | 14,717 | +6% | 1 | 1 | 0% | 2,250 | 2,912 | +29% | 0 | 0 | — |
case-06 | pass→pass | 12,531 | 12,186 | -3% | 1 | 1 | 0% | 2,348 | 2,476 | +5% | 0 | 0 | — |
case-07 | pass→pass | 16,730 | 10,277 | -39% | 1 | 1 | 0% | 2,351 | 2,493 | +6% | 0 | 0 | — |
case-08 | pass→pass | 11,589 | 11,093 | -4% | 1 | 1 | 0% | 2,166 | 2,286 | +6% | 0 | 0 | — |
case-09 | pass→pass | 11,118 | 10,198 | -8% | 1 | 1 | 0% | 2,006 | 2,244 | +12% | 0 | 0 | — |
case-10 | pass→pass | 11,344 | 10,832 | -5% | 1 | 1 | 0% | 2,259 | 2,310 | +2% | 0 | 0 | — |
case-11 | fail→pass | 12,363 | 11,236 | -9% | 1 | 1 | 0% | 2,163 | 2,456 | +14% | 0 | 0 | — |
case-12 | fail→pass | 15,100 | 15,218 | +1% | 1 | 1 | 0% | 2,832 | 3,131 | +11% | 0 | 0 | — |
case-13 | pass→pass | 14,919 | 14,063 | -6% | 1 | 1 | 0% | 2,396 | 2,664 | +11% | 0 | 0 | — |
case-14 | pass→pass | 14,850 | 14,930 | +1% | 1 | 1 | 0% | 2,708 | 2,505 | -7% | 0 | 0 | — |
case-15 | pass→pass | 14,124 | 14,253 | +1% | 1 | 1 | 0% | 2,289 | 2,455 | +7% | 0 | 0 | — |
case-16 | pass→pass | 11,950 | 12,638 | +6% | 1 | 1 | 0% | 1,916 | 2,180 | +14% | 0 | 0 | — |
case-17 | pass→pass | 13,262 | 9,997 | -25% | 1 | 1 | 0% | 2,386 | 1,991 | -17% | 0 | 0 | — |
case-18 | pass→pass | 19,029 | 15,328 | -19% | 1 | 1 | 0% | 2,763 | 2,900 | +5% | 0 | 0 | — |
case-19 | pass→pass | 10,223 | 11,665 | +14% | 1 | 1 | 0% | 1,802 | 2,261 | +25% | 0 | 0 | — |
case-20 | pass→pass | 19,405 | 15,126 | -22% | 1 | 1 | 0% | 2,792 | 3,220 | +15% | 0 | 0 | — |
case-21 | fail→pass | 17,460 | 13,917 | -20% | 1 | 1 | 0% | 2,716 | 3,041 | +12% | 0 | 0 | — |
case-22 | pass→pass | 14,603 | 14,361 | -2% | 1 | 1 | 0% | 2,830 | 2,479 | -12% | 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. 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.