Loading skill
Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Expert skill in battery pack design for automotive EVs, covering cell selection, module layout, thermal management, and safety analysis. Covers 5 topics across ev-tools domain. Includes 5 skill files covering ASPICE Level 2/3 for software process, ASPICE Level 3, AUTOSAR 4.4, AUTOSAR 4.4 for software architecture, IEC 61851 EV charging systems, IEC 63110 Management of EV charging infrastructure, IPC-2221 PCB design standards, ISO 12405 Lithium-ion battery testing and more.
.claude/skills/pangzhenying2025-automotive-ev-tools/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-09 | ✗→✓ | ▲ Improved | 266% | 0% |
| case-21 | ✗→✓ | ▲ Improved | 735% | 0% |
| case-04 | ✓→✓ | ＝ Same ✓ | 322% | 0% |
| case-05 | ✓→✓ | ＝ Same ✓ | 537% | 0% |
| case-06 | ✓→✓ | ＝ Same ✓ | 238% | 0% |
5 skill files covering ev-tools domain for automotive software engineering.
Expert in automotive battery pack design from requirements definition through production validation, including cell selection, module layout, thermal management, safety analysis, and cost optimization.
Expert in LibreSolar modular open-source battery management system, covering hardware design, firmware development, and system integration for automotive EV and stationary energy storage.
/battery/voltage, /battery/current, /battery/soc/cells/voltages, /cells/temperatures/config/limits/voltage_max, /config/limits/current_charge_maxExpert in Open Charge Point Protocol (OCPP) implementation for electric vehicle charging infrastructure, covering charge point firmware, central system integration, and smart charging orchestration.
json{ "chargingProfileId": 1, "stackLevel": 0, "chargingProfilePurpose": "ChargePointMaxProfile", "chargingProfileKind": "Recurring", "recurrencyKind": "Daily", "chargingSchedule": { "startSchedule": "2026-03-19T18:00:00Z", "duration": 14400, "chargingRateUnit": "W", "chargingSchedulePeriod": [ {"startPeriod": 0, "limit": 11000} ] } }
Expert in OpenBMS open-source battery management system architecture, algorithms, and integration for automotive lithium-ion battery packs.
Expert in physics-based battery modeling using PyBaMM (Python Battery Mathematical Modeling) framework for automotive lithium-ion battery systems.
| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-01 | fail→fail | 33,179 | 32,612 | -2% | 1 | 1 | 0% | 6,211 | 14,019 | +126% | 0 | 0 | — |
case-02 | fail→fail | 30,271 | 29,989 | -1% | 1 | 1 | 0% | 6,217 | 14,025 | +126% | 0 | 0 | — |
case-03 | fail→fail | 42,860 | 17,885 | -58% | 1 | 1 | 0% | 1,744 | 11,145 | +539% | 0 | 0 | — |
case-04 | pass→pass | 10,370 | 9,194 | -11% | 1 | 1 | 0% | 2,306 | 9,733 | +322% | 0 | 0 | — |
case-05 | pass→pass | 7,180 | 7,252 | +1% | 1 | 1 | 0% | 1,449 | 9,234 | +537% | 0 | 0 | — |
case-06 | pass→pass | 14,432 | 8,475 | -41% | 1 | 1 | 0% | 2,821 | 9,530 | +238% | 0 | 0 | — |
case-07 | pass→pass | 13,324 | 6,441 | -52% | 1 | 1 | 0% | 2,329 | 9,058 | +289% | 0 | 0 | — |
case-08 | pass→pass | 15,547 | 10,537 | -32% | 1 | 1 | 0% | 2,718 | 9,698 | +257% | 0 | 0 | — |
case-09 | fail→pass | 12,415 | 11,068 | -11% | 1 | 1 | 0% | 2,679 | 9,800 | +266% | 0 | 0 | — |
case-10 | pass→pass | 4,985 | 6,834 | +37% | 1 | 1 | 0% | 773 | 8,943 | +1057% | 0 | 0 | — |
case-11 | pass→pass | 18,784 | 18,425 | -2% | 1 | 1 | 0% | 3,459 | 11,356 | +228% | 0 | 0 | — |
case-12 | pass→pass | 10,404 | 9,057 | -13% | 1 | 1 | 0% | 2,007 | 9,591 | +378% | 0 | 0 | — |
case-13 | pass→pass | 10,122 | 11,677 | +15% | 1 | 1 | 0% | 2,137 | 10,204 | +377% | 0 | 0 | — |
case-14 | pass→pass | 18,395 | 18,418 | +0% | 1 | 1 | 0% | 3,259 | 11,181 | +243% | 0 | 0 | — |
case-15 | pass→pass | 15,463 | 12,941 | -16% | 1 | 1 | 0% | 2,749 | 10,180 | +270% | 0 | 0 | — |
case-16 | fail→fail | 13,114 | 8,569 | -35% | 1 | 1 | 0% | 2,488 | 9,412 | +278% | 0 | 0 | — |
case-17 | pass→pass | 7,510 | 5,480 | -27% | 1 | 1 | 0% | 1,453 | 8,752 | +502% | 0 | 0 | — |
case-18 | pass→pass | 13,695 | 5,892 | -57% | 1 | 1 | 0% | 2,579 | 8,860 | +244% | 0 | 0 | — |
case-19 | pass→pass | 6,876 | 6,020 | -12% | 1 | 1 | 0% | 1,261 | 8,724 | +592% | 0 | 0 | — |
case-20 | pass→pass | 27,207 | 26,807 | -1% | 1 | 1 | 0% | 6,001 | 14,014 | +134% | 0 | 0 | — |
case-21 | fail→pass | 44,824 | 23,312 | -48% | 1 | 1 | 0% | 1,512 | 12,621 | +735% | 0 | 0 | — |
case-22 | pass→pass | 35,104 | 31,246 | -11% | 1 | 1 | 0% | 6,037 | 14,011 | +132% | 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, and 20 counted toward the lift figure. The other 2 produced results that are not comparable between the two arms, so they are excluded from the headline rather than averaged into it. The headline lift of +9 percentage points is the difference between those two pass rates over the 20 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.