Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Cursor rules for Astro development with TypeScript integration.
.claude/skills/amariahak-cursor-rules-for-astro-development-with-typescript-integration/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-03 | ✗→✓ | ▲ Improved | -20% | 0% |
| case-05 | ✗→✓ | ▲ Improved | 131% | 0% |
| case-06 | ✗→✓ | ▲ Improved | -20% | 0% |
| case-08 | ✗→✓ | ▲ Improved | 105% | 0% |
| case-09 | ✗→✓ | ▲ Improved | 7% | 0% |
Cursor rules for Astro development with TypeScript integration.
Synced from https://github.com/PatrickJS/awesome-cursorrules/tree/main/rules/astro-typescript-cursorrules-prompt-file.mdc.
{ "rules": { "commit_message_guidelines": { "description": "Guidelines for creating conventional commit messages.",
"format": { "description": "The format for commit messages using the conventional commits spec.", "body": "optional scope]: \n\noptional body]\n\noptional footer(s)]" },
"enabled": true,
"rules": { "description": "Always suggest a conventional commit with a type and optional scope in lowercase letters." }, { "description": "Keep the commit message concise and within 60 characters." }, { "description": "Ensure the commit message is ready to be pasted into the terminal without further editing." }, { "description": "Provide the full command to commit, not just the message." } ],
"examples": { "prompt": "<diff_context> /commit", "response": "git commit -m 'feat: add responsive navbar with TailwindCSS'" } ] },
"development_guidelines": { "description": "Guidelines for developing code with Astro, TypeScript, and TailwindCSS.",
"enabled": true,
"rules": { "description": "Enforce strict TypeScript settings, ensuring type safety across the project." }, { "description": "Use TailwindCSS for all styling, keeping the utility-first approach in mind." }, { "description": "Ensure Astro components are modular, reusable, and maintain a clear separation of concerns." } ] },
"coding_style": { "description": "Guidelines for maintaining consistent coding style.",
"enabled": true,
"rules": { "description": "Code must start with path/filename as a one-line comment." }, { "description": "Comments should describe purpose, not effect." }, { "description": "Prioritize modularity, DRY principles, and performance." } ] },
"custom_slash_commands": { "description": "Custom slash commands.",
"enabled": true,
"commands": { "name": "/commit", "description": "Generate a Git commit message using the conventional commits spec.", "enabled": true } ] } } }
| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-01 | fail→fail | 15,146 | 8,557 | -44% | 1 | 1 | 0% | 3,589 | 2,630 | -27% | 0 | 0 | — |
case-02 | pass→pass | 2,644 | 2,370 | -10% | 1 | 1 | 0% | 444 | 1,022 | +130% | 0 | 0 | — |
case-03 | fail→pass | 15,188 | 9,109 | -40% | 1 | 1 | 0% | 3,397 | 2,702 | -20% | 0 | 0 | — |
case-04 | fail→fail | 16,249 | 6,562 | -60% | 1 | 1 | 0% | 3,887 | 2,111 | -46% | 0 | 0 | — |
case-05 | fail→pass | 3,324 | 2,801 | -16% | 1 | 1 | 0% | 491 | 1,135 | +131% | 0 | 0 | — |
case-06 | fail→pass | 17,840 | 10,885 | -39% | 1 | 1 | 0% | 4,138 | 3,299 | -20% | 0 | 0 | — |
case-07 | fail→fail | 15,461 | 11,369 | -26% | 1 | 1 | 0% | 3,187 | 3,243 | +2% | 0 | 0 | — |
case-08 | fail→pass | 2,976 | 2,594 | -13% | 1 | 1 | 0% | 547 | 1,124 | +105% | 0 | 0 | — |
case-09 | fail→pass | 11,277 | 8,812 | -22% | 1 | 1 | 0% | 2,478 | 2,647 | +7% | 0 | 0 | — |
case-10 | fail→pass | 4,668 | 2,749 | -41% | 1 | 1 | 0% | 832 | 1,140 | +37% | 0 | 0 | — |
case-11 | fail→pass | 10,625 | 5,459 | -49% | 1 | 1 | 0% | 2,179 | 1,648 | -24% | 0 | 0 | — |
case-12 | fail→fail | 19,299 | 21,006 | +9% | 1 | 1 | 0% | 4,574 | 4,569 | -0% | 0 | 0 | — |
case-13 | fail→pass | 5,055 | 2,094 | -59% | 1 | 1 | 0% | 861 | 1,018 | +18% | 0 | 0 | — |
case-14 | fail→pass | 16,176 | 8,763 | -46% | 1 | 1 | 0% | 3,516 | 2,667 | -24% | 0 | 0 | — |
case-15 | fail→fail | 14,323 | 10,916 | -24% | 1 | 1 | 0% | 2,898 | 3,021 | +4% | 0 | 0 | — |
case-16 | pass→pass | 4,510 | 2,184 | -52% | 1 | 1 | 0% | 803 | 997 | +24% | 0 | 0 | — |
case-17 | fail→fail | 20,801 | 17,040 | -18% | 1 | 1 | 0% | 4,459 | 4,824 | +8% | 0 | 0 | — |
case-18 | fail→fail | 9,413 | 4,789 | -49% | 1 | 1 | 0% | 1,854 | 1,667 | -10% | 0 | 0 | — |
case-19 | fail→fail | 9,181 | 5,813 | -37% | 1 | 1 | 0% | 1,917 | 1,800 | -6% | 0 | 0 | — |
case-20 | pass→pass | 14,877 | 13,420 | -10% | 1 | 1 | 0% | 2,548 | 3,000 | +18% | 0 | 0 | — |
case-21 | pass→pass | 15,023 | 9,984 | -34% | 1 | 1 | 0% | 2,490 | 2,267 | -9% | 0 | 0 | — |
case-22 | pass→fail | 11,914 | 9,170 | -23% | 1 | 1 | 0% | 1,866 | 2,157 | +16% | 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 +36 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.