Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Figure out whether your problem car might qualify for a refund or replacement under lemon law or warranty — and build the paper trail to claim it. Use when asked is my car a lemon, my new car keeps breaking, lemon law help, or can I return a defective car. Produces a plausibility read against typical lemon-law criteria (repeated same defect, repair attempts, time out of service, warranty window), the records to gather, the manufacturer-claim and escalation steps, and a strong flag that lemon law
.claude/skills/mohitagw15856-lemon-law-check/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-10 | ✗→✓ | ▲ Improved | 12% | 0% |
| case-17 | ✗→✓ | ▲ Improved | 27% | 0% |
| case-04 | ✓→✓ | = Same ✓ | 19% | 0% |
| case-05 | ✓→✓ | = Same ✓ | 38% | 0% |
| case-06 | ✓→✓ | = Same ✓ | 33% | 0% |
A car that keeps failing for the same fault, despite repeated repair attempts, may qualify for a refund or replacement under lemon laws or warranty — but only if you meet the criteria and have the documentation. This checks whether your situation plausibly fits, tells you exactly what records to keep, and lays out the manufacturer claim and escalation — while being clear that lemon laws vary by jurisdiction and this isn't legal advice.
Ask for these if not provided:
Plausible lemon? vs typical criteria: same defect n] attempts / days] out of service / within warranty → likely / borderline / probably not]. Gather: every repair order (complaint + fix) · dates · days in shop · all communications. Claim: notify manufacturer in writing → allow required final repair → request refund/replacement. Escalate: arbitration → consumer authority → lemon-law attorney].
> Lemon laws are jurisdiction-specific and this isn't legal advice. Confirm your local criteria; a lemon-law attorney is often free to you if the manufacturer pays fees on a win.
| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-01 | fail→fail | 18,109 | 20,463 | +13% | 1 | 1 | 0% | 2,940 | 3,194 | +9% | 0 | 0 | — |
case-02 | fail→fail | 25,345 | 16,295 | -36% | 1 | 1 | 0% | 3,117 | 3,891 | +25% | 0 | 0 | — |
case-03 | fail→fail | 27,181 | 19,212 | -29% | 1 | 1 | 0% | 3,155 | 3,236 | +3% | 0 | 0 | — |
case-04 | pass→pass | 21,433 | 19,952 | -7% | 1 | 1 | 0% | 2,648 | 3,152 | +19% | 0 | 0 | — |
case-05 | pass→pass | 19,829 | 13,964 | -30% | 1 | 1 | 0% | 2,370 | 3,265 | +38% | 0 | 0 | — |
case-06 | pass→pass | 16,968 | 21,578 | +27% | 1 | 1 | 0% | 2,781 | 3,691 | +33% | 0 | 0 | — |
case-07 | pass→pass | 16,813 | 18,984 | +13% | 1 | 1 | 0% | 2,713 | 3,125 | +15% | 0 | 0 | — |
case-08 | pass→pass | 16,827 | 13,470 | -20% | 1 | 1 | 0% | 1,914 | 2,445 | +28% | 0 | 0 | — |
case-09 | pass→pass | 11,684 | 12,189 | +4% | 1 | 1 | 0% | 1,897 | 2,250 | +19% | 0 | 0 | — |
case-10 | fail→pass | 17,589 | 14,045 | -20% | 1 | 1 | 0% | 2,051 | 2,289 | +12% | 0 | 0 | — |
case-11 | pass→pass | 18,571 | 16,811 | -9% | 1 | 1 | 0% | 2,095 | 2,821 | +35% | 0 | 0 | — |
case-12 | pass→pass | 18,204 | 11,649 | -36% | 1 | 1 | 0% | 2,039 | 2,834 | +39% | 0 | 0 | — |
case-13 | pass→pass | 20,486 | 16,790 | -18% | 1 | 1 | 0% | 2,390 | 2,948 | +23% | 0 | 0 | — |
case-14 | fail→fail | 13,559 | 14,186 | +5% | 1 | 1 | 0% | 2,202 | 2,926 | +33% | 0 | 0 | — |
case-15 | pass→pass | 17,403 | 11,846 | -32% | 1 | 1 | 0% | 2,075 | 2,929 | +41% | 0 | 0 | — |
case-16 | pass→pass | 19,084 | 18,495 | -3% | 1 | 1 | 0% | 2,515 | 2,922 | +16% | 0 | 0 | — |
case-17 | fail→pass | 17,558 | 15,276 | -13% | 1 | 1 | 0% | 2,267 | 2,884 | +27% | 0 | 0 | — |
case-18 | pass→pass | 13,097 | 8,664 | -34% | 1 | 1 | 0% | 2,007 | 2,561 | +28% | 0 | 0 | — |
case-19 | pass→pass | 16,743 | 16,288 | -3% | 1 | 1 | 0% | 1,874 | 2,848 | +52% | 0 | 0 | — |
case-20 | pass→pass | 16,813 | 18,677 | +11% | 1 | 1 | 0% | 2,050 | 2,887 | +41% | 0 | 0 | — |
case-21 | fail→fail | 13,240 | 14,319 | +8% | 1 | 1 | 0% | 2,348 | 3,264 | +39% | 0 | 0 | — |
case-22 | pass→pass | 17,367 | 11,193 | -36% | 1 | 1 | 0% | 2,092 | 2,651 | +27% | 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.
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.