Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Cursor rules for Laravel development with PHP 8.3 integration.
.claude/skills/amariahak-cursor-rules-for-laravel-development-with-php-8/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-10 | ✗→✓ | ▲ Improved | 11% | 0% |
| case-01 | ✗→✓ | ▲ Improved | -16% | 0% |
| case-02 | ✗→✓ | ▲ Improved | 2% | 0% |
| case-03 | ✗→✓ | ▲ Improved | 35% | 0% |
| case-12 | ✗→✓ | ▲ Improved | -8% | 0% |
Cursor rules for Laravel development with PHP 8.3 integration.
Synced from https://github.com/PatrickJS/awesome-cursorrules/tree/main/rules/laravel-php-83-cursorrules-prompt-file.mdc.
You are a highly skilled Laravel package developer tasked with creating a new package. Your goal is to provide a detailed plan and code structure for the package based on the given project description and specific requirements.
Remember to adhere to the specified coding standards, development guidelines, and Laravel best practices throughout your plan and code samples. Ensure that your response is detailed, well-structured, and provides a clear roadmap for developing the Laravel package based on the given project description and requirements.
| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-10 | fail→pass | 15,690 | 17,018 | +8% | 1 | 1 | 0% | 3,473 | 3,853 | +11% | 0 | 0 | — |
case-04 | pass→fail | 16,088 | 14,820 | -8% | 1 | 1 | 0% | 3,261 | 3,647 | +12% | 0 | 0 | — |
case-05 | pass→pass | 17,046 | 14,639 | -14% | 1 | 1 | 0% | 3,264 | 3,342 | +2% | 0 | 0 | — |
case-01 | fail→pass | 28,489 | 22,074 | -23% | 1 | 1 | 0% | 6,208 | 5,215 | -16% | 0 | 0 | — |
case-02 | fail→pass | 28,215 | 31,426 | +11% | 1 | 1 | 0% | 5,694 | 5,788 | +2% | 0 | 0 | — |
case-03 | fail→pass | 22,964 | 22,389 | -3% | 1 | 1 | 0% | 4,022 | 5,411 | +35% | 0 | 0 | — |
case-06 | pass→pass | 15,423 | 13,291 | -14% | 1 | 1 | 0% | 2,763 | 2,897 | +5% | 0 | 0 | — |
case-07 | fail→fail | 15,190 | 18,870 | +24% | 1 | 1 | 0% | 3,588 | 4,733 | +32% | 0 | 0 | — |
case-08 | fail→fail | 15,440 | 14,167 | -8% | 1 | 1 | 0% | 3,073 | 3,496 | +14% | 0 | 0 | — |
case-09 | fail→fail | 16,639 | 15,156 | -9% | 1 | 1 | 0% | 3,698 | 3,601 | -3% | 0 | 0 | — |
case-11 | fail→fail | 14,399 | 14,053 | -2% | 1 | 1 | 0% | 3,176 | 3,512 | +11% | 0 | 0 | — |
case-12 | fail→pass | 15,687 | 12,078 | -23% | 1 | 1 | 0% | 3,223 | 2,955 | -8% | 0 | 0 | — |
case-13 | fail→fail | 13,015 | 17,010 | +31% | 1 | 1 | 0% | 2,428 | 4,142 | +71% | 0 | 0 | — |
case-14 | fail→fail | 21,094 | 15,386 | -27% | 1 | 1 | 0% | 3,391 | 3,881 | +14% | 0 | 0 | — |
case-15 | fail→fail | 13,517 | 16,707 | +24% | 1 | 1 | 0% | 2,655 | 3,988 | +50% | 0 | 0 | — |
case-16 | pass→pass | 10,317 | 11,359 | +10% | 1 | 1 | 0% | 1,995 | 2,962 | +48% | 0 | 0 | — |
case-17 | fail→fail | 10,925 | 12,873 | +18% | 1 | 1 | 0% | 2,299 | 3,110 | +35% | 0 | 0 | — |
case-18 | fail→fail | 10,288 | 6,745 | -34% | 1 | 1 | 0% | 2,112 | 1,938 | -8% | 0 | 0 | — |
case-19 | fail→fail | 13,669 | 10,645 | -22% | 1 | 1 | 0% | 2,638 | 2,555 | -3% | 0 | 0 | — |
case-20 | fail→fail | 12,503 | 10,561 | -16% | 1 | 1 | 0% | 2,664 | 2,571 | -3% | 0 | 0 | — |
case-21 | fail→pass | 13,221 | 13,147 | -1% | 1 | 1 | 0% | 2,941 | 3,196 | +9% | 0 | 0 | — |
case-22 | fail→fail | 9,848 | 16,145 | +64% | 1 | 1 | 0% | 2,234 | 3,325 | +49% | 0 | 0 | — |
case-23 | fail→fail | 14,047 | 19,188 | +37% | 1 | 1 | 0% | 2,903 | 4,761 | +64% | 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. 23 cases were attempted. The headline lift of +22 percentage points is the difference between those two pass rates over the 23 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.