Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Project structure and file organization guidelines
.claude/skills/elie222-project-structure/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-02 | ✗→✓ | ▲ Improved | 11% | 0% |
| case-03 | ✗→✓ | ▲ Improved | -21% | 0% |
| case-04 | ✗→✓ | ▲ Improved | 12% | 0% |
| case-05 | ✗→✓ | ▲ Improved | 1% | 0% |
| case-06 | ✗→✓ | ▲ Improved | -14% | 0% |
apps/webpackages folderapps/web/utils/actions foldertree. ├── apps │ ├── web/ # Main Next.js application │ │ ├── app/ # Next.js App Router │ │ │ ├── (app)/ # Main application pages │ │ │ │ ├── assistant/ # AI assistant feature │ │ │ │ ├── reply-zero/ # Reply Zero feature │ │ │ │ ├── settings/ # User settings │ │ │ │ ├── setup/ # Main onboarding │ │ │ │ ├── clean/ # Bulk email cleanup │ │ │ │ ├── smart-categories/ # Smart sender categorization │ │ │ │ ├── bulk-unsubscribe/ # Bulk unsubscribe │ │ │ │ ├── stats/ # Email analytics │ │ │ │ ├── mail/ # Email client (in beta) │ │ │ │ └── ... (other app routes) │ │ │ ├── api/ # API Routes │ │ │ │ ├── knowledge/ # Knowledge base API │ │ │ │ ├── reply-tracker/ # Reply tracking │ │ │ │ ├── clean/ # Cleanup API │ │ │ │ ├── ai/ # AI features API │ │ │ │ ├── user/ # User management │ │ │ │ ├── google/ # Google integration │ │ │ │ ├── auth/ # Authentication │ │ │ │ └── ... (other APIs) │ │ │ ├── (landing)/ # Marketing/landing pages │ │ │ ├── blog/ # Blog pages │ │ │ ├── layout.tsx # Root layout │ │ │ └── ... (other app files) │ │ ├── utils/ # Utility functions and helpers │ │ │ ├── actions/ # Server actions │ │ │ ├── ai/ # AI-related utilities │ │ │ ├── llms/ # Language model utilities │ │ │ ├── gmail/ # Gmail integration utilities │ │ │ ├── redis/ # Redis utilities │ │ │ ├── user/ # User-related utilities │ │ │ ├── parse/ # Parsing utilities │ │ │ ├── queue/ # Queue management │ │ │ ├── error-messages/ # Error handling │ │ │ └── *.ts # Other utility files (auth, email, etc.) │ │ ├── public/ # Static assets (images, fonts) │ │ ├── prisma/ # Prisma schema and client │ │ ├── styles/ # Global CSS styles │ │ ├── providers/ # React Context providers │ │ ├── hooks/ # Custom React hooks │ │ ├── sanity/ # Sanity CMS integration │ │ ├── __tests__/ # AI test files (Vitest) │ │ ├── scripts/ # Utility scripts │ │ ├── store/ # State management │ │ ├── types/ # TypeScript type definitions │ │ ├── next.config.mjs │ │ ├── package.json │ │ └── ... (config files) ├── packages ├── tinybird/ ├── loops/ ├── resend/ ├── tinybird-ai-analytics/ └── tsconfig/
api/hello-world/route)components/Button.tsx)components/uicomponents/components folderapps/web/app/(app)/PAGE_NAME/page.tsxpage.tsx or in the apps/web/app/(app)/PAGE_NAME folderswr for data fetching in deeply nested componentsonClick must be client components with use client directiveuse serverutils/ folder for reusable logicts import groupBy from "lodash/groupBy";
| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-09 | pass→pass | 7,088 | 2,612 | -63% | 1 | 1 | 0% | 1,284 | 1,552 | +21% | 0 | 0 | — |
case-01 | fail→fail | 17,109 | 12,158 | -29% | 1 | 1 | 0% | 3,417 | 3,012 | -12% | 0 | 0 | — |
case-02 | fail→pass | 11,496 | 5,557 | -52% | 1 | 1 | 0% | 1,974 | 2,191 | +11% | 0 | 0 | — |
case-03 | fail→pass | 16,162 | 5,574 | -66% | 1 | 1 | 0% | 2,872 | 2,268 | -21% | 0 | 0 | — |
case-04 | fail→pass | 8,930 | 3,262 | -63% | 1 | 1 | 0% | 1,575 | 1,761 | +12% | 0 | 0 | — |
case-05 | fail→pass | 11,061 | 3,614 | -67% | 1 | 1 | 0% | 1,788 | 1,810 | +1% | 0 | 0 | — |
case-06 | fail→pass | 14,870 | 5,272 | -65% | 1 | 1 | 0% | 2,423 | 2,080 | -14% | 0 | 0 | — |
case-07 | pass→pass | 6,269 | 3,462 | -45% | 1 | 1 | 0% | 1,092 | 1,707 | +56% | 0 | 0 | — |
case-08 | fail→pass | 6,144 | 2,371 | -61% | 1 | 1 | 0% | 1,045 | 1,547 | +48% | 0 | 0 | — |
case-10 | pass→pass | 6,136 | 2,260 | -63% | 1 | 1 | 0% | 1,120 | 1,490 | +33% | 0 | 0 | — |
case-11 | pass→pass | 10,484 | 3,880 | -63% | 1 | 1 | 0% | 1,695 | 1,870 | +10% | 0 | 0 | — |
case-12 | pass→pass | 10,346 | 3,855 | -63% | 1 | 1 | 0% | 1,676 | 1,677 | +0% | 0 | 0 | — |
case-13 | fail→pass | 13,782 | 2,408 | -83% | 1 | 1 | 0% | 2,355 | 1,470 | -38% | 0 | 0 | — |
case-14 | fail→pass | 6,648 | 3,953 | -41% | 1 | 1 | 0% | 1,092 | 1,641 | +50% | 0 | 0 | — |
case-15 | fail→pass | 6,827 | 2,552 | -63% | 1 | 1 | 0% | 1,102 | 1,622 | +47% | 0 | 0 | — |
case-16 | pass→pass | 13,179 | 3,595 | -73% | 1 | 1 | 0% | 2,386 | 1,607 | -33% | 0 | 0 | — |
case-17 | fail→pass | 11,146 | 2,591 | -77% | 1 | 1 | 0% | 2,108 | 1,584 | -25% | 0 | 0 | — |
case-18 | fail→pass | 8,297 | 2,338 | -72% | 1 | 1 | 0% | 1,194 | 1,541 | +29% | 0 | 0 | — |
case-19 | fail→pass | 13,197 | 1,889 | -86% | 1 | 1 | 0% | 2,001 | 1,396 | -30% | 0 | 0 | — |
case-20 | pass→pass | 7,188 | 5,390 | -25% | 1 | 1 | 0% | 1,120 | 2,026 | +81% | 0 | 0 | — |
case-21 | pass→pass | 6,994 | 2,533 | -64% | 1 | 1 | 0% | 1,254 | 1,582 | +26% | 0 | 0 | — |
case-22 | pass→pass | 8,952 | 4,364 | -51% | 1 | 1 | 0% | 1,371 | 1,864 | +36% | 0 | 0 | — |
case-23 | fail→pass | 7,940 | 2,175 | -73% | 1 | 1 | 0% | 1,383 | 1,481 | +7% | 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 +57 percentage points is the difference between those two pass rates over the 23 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.