Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Cursor rules for HTMX development with Django integration.
.claude/skills/amariahak-cursor-rules-for-htmx-development-with-django-integration/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-17 | ✗→✓ | ▲ Improved | 43% | 0% |
| case-20 | ✗→✓ | ▲ Improved | -14% | 0% |
| case-07 | ✓→✗ | ▼ Worse | -16% | 0% |
| case-03 | ✓→✓ | = Same ✓ | -9% | 0% |
| case-04 | ✓→✓ | = Same ✓ | 2% | 0% |
Cursor rules for HTMX development with Django integration.
Synced from https://github.com/PatrickJS/awesome-cursorrules/tree/main/rules/htmx-django-cursorrules-prompt-file.mdc.
// HTMX with Django .cursorrules
// HTMX and Django best practices
const htmxDjangoBestPractices = "Use Django's template system with HTMX attributes", "Implement Django forms for form handling", "Utilize Django's URL routing system", "Use Django's class-based views for HTMX responses", "Implement Django ORM for database operations", "Utilize Django's middleware for request/response processing", ];
// Folder structure
const folderStructure = project_name/ app_name/ templates/ static/ css/ js/ models.py views.py urls.py project_name/ settings.py urls.py manage.py ;
// Additional instructions
const additionalInstructions =
;
| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-01 | fail→fail | 18,142 | 12,410 | -32% | 1 | 1 | 0% | 3,602 | 2,961 | -18% | 0 | 0 | — |
case-02 | fail→fail | 13,764 | 12,112 | -12% | 1 | 1 | 0% | 2,732 | 2,875 | +5% | 0 | 0 | — |
case-03 | pass→pass | 12,116 | 9,226 | -24% | 1 | 1 | 0% | 2,287 | 2,087 | -9% | 0 | 0 | — |
case-04 | pass→pass | 11,284 | 10,425 | -8% | 1 | 1 | 0% | 2,103 | 2,143 | +2% | 0 | 0 | — |
case-05 | pass→pass | 16,936 | 16,396 | -3% | 1 | 1 | 0% | 3,465 | 3,683 | +6% | 0 | 0 | — |
case-06 | pass→pass | 12,246 | 11,472 | -6% | 1 | 1 | 0% | 2,151 | 2,593 | +21% | 0 | 0 | — |
case-07 | pass→fail | 16,049 | 11,268 | -30% | 1 | 1 | 0% | 2,968 | 2,487 | -16% | 0 | 0 | — |
case-08 | pass→pass | 11,006 | 6,744 | -39% | 1 | 1 | 0% | 1,776 | 1,442 | -19% | 0 | 0 | — |
case-09 | pass→pass | 11,141 | 6,791 | -39% | 1 | 1 | 0% | 1,930 | 1,575 | -18% | 0 | 0 | — |
case-10 | pass→pass | 12,721 | 10,907 | -14% | 1 | 1 | 0% | 2,361 | 2,328 | -1% | 0 | 0 | — |
case-11 | pass→pass | 7,879 | 4,228 | -46% | 1 | 1 | 0% | 1,531 | 1,111 | -27% | 0 | 0 | — |
case-12 | pass→pass | 15,912 | 13,999 | -12% | 1 | 1 | 0% | 2,774 | 2,883 | +4% | 0 | 0 | — |
case-13 | pass→pass | 6,423 | 5,575 | -13% | 1 | 1 | 0% | 1,162 | 1,154 | -1% | 0 | 0 | — |
case-14 | pass→pass | 18,342 | 10,148 | -45% | 1 | 1 | 0% | 3,326 | 2,243 | -33% | 0 | 0 | — |
case-15 | pass→pass | 15,021 | 14,782 | -2% | 1 | 1 | 0% | 2,585 | 2,932 | +13% | 0 | 0 | — |
case-16 | pass→pass | 10,349 | 4,555 | -56% | 1 | 1 | 0% | 2,005 | 1,177 | -41% | 0 | 0 | — |
case-17 | fail→pass | 6,305 | 6,800 | +8% | 1 | 1 | 0% | 1,170 | 1,675 | +43% | 0 | 0 | — |
case-18 | pass→pass | 15,413 | 12,856 | -17% | 1 | 1 | 0% | 2,818 | 2,918 | +4% | 0 | 0 | — |
case-19 | pass→pass | 3,239 | 2,301 | -29% | 1 | 1 | 0% | 548 | 724 | +32% | 0 | 0 | — |
case-20 | fail→pass | 5,624 | 3,182 | -43% | 1 | 1 | 0% | 979 | 843 | -14% | 0 | 0 | — |
case-21 | pass→pass | 6,268 | 4,460 | -29% | 1 | 1 | 0% | 1,116 | 956 | -14% | 0 | 0 | — |
case-22 | pass→pass | 9,783 | 6,575 | -33% | 1 | 1 | 0% | 1,749 | 1,645 | -6% | 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 +5 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.