Loading skill
Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Interaction with NeonDB Postgres using Drizzle ORM.
.claude/skills/aiskillstore-database-orm/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-01 | ✗→✓ | ▲ Improved | -53% | 0% |
| case-02 | ✗→✓ | ▲ Improved | -55% | 0% |
| case-04 | ✗→✓ | ▲ Improved | -65% | 0% |
| case-06 | ✗→✓ | ▲ Improved | -14% | 0% |
| case-07 | ✗→✓ | ▲ Improved | -44% | 0% |
@neondatabase/serverlessThe database connection is initialized in db/index.ts.
tsimport { neon } from '@neondatabase/serverless'; import { drizzle } from 'drizzle-orm/neon-http'; const sql = neon(process.env.DATABASE_URL!); export const db = drizzle(sql);
Schema definitions are in db/schema.ts.
users, sessions, accounts, verifications: Auth tables.analyses, chatbot_history: App specific tables.Example of a database query:
tsimport { db } from "@/db"; import { users } from "@/db/schema"; import { eq } from "drizzle-orm"; // Select const user = await db.select().from(users).where(eq(users.email, "test@example.com")); // Insert await db.insert(users).values({ ... });
npx drizzle-kit generatenpx drizzle-kit push (or migrate script)| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-01 | fail→pass | 12,856 | 4,048 | -69% | 1 | 1 | 0% | 2,140 | 1,007 | -53% | 0 | 0 | — |
case-02 | fail→pass | 9,785 | 3,033 | -69% | 1 | 1 | 0% | 1,727 | 771 | -55% | 0 | 0 | — |
case-03 | pass→pass | 15,481 | 4,750 | -69% | 1 | 1 | 0% | 2,627 | 1,132 | -57% | 0 | 0 | — |
case-04 | fail→pass | 14,051 | 3,316 | -76% | 1 | 1 | 0% | 2,339 | 810 | -65% | 0 | 0 | — |
case-05 | pass→pass | 9,572 | 1,829 | -81% | 1 | 1 | 0% | 1,550 | 547 | -65% | 0 | 0 | — |
case-06 | fail→pass | 6,719 | 4,246 | -37% | 1 | 1 | 0% | 1,297 | 1,121 | -14% | 0 | 0 | — |
case-07 | fail→pass | 8,040 | 3,166 | -61% | 1 | 1 | 0% | 1,472 | 829 | -44% | 0 | 0 | — |
case-08 | pass→pass | 6,906 | 1,488 | -78% | 1 | 1 | 0% | 1,283 | 522 | -59% | 0 | 0 | — |
case-09 | pass→pass | 8,477 | 2,205 | -74% | 1 | 1 | 0% | 1,476 | 652 | -56% | 0 | 0 | — |
case-10 | pass→pass | 6,368 | 1,264 | -80% | 1 | 1 | 0% | 1,101 | 464 | -58% | 0 | 0 | — |
case-11 | fail→pass | 2,560 | 1,694 | -34% | 1 | 1 | 0% | 294 | 528 | +80% | 0 | 0 | — |
case-12 | pass→pass | 6,380 | 1,454 | -77% | 1 | 1 | 0% | 1,083 | 489 | -55% | 0 | 0 | — |
case-13 | pass→pass | 5,999 | 1,933 | -68% | 1 | 1 | 0% | 995 | 601 | -40% | 0 | 0 | — |
case-14 | pass→pass | 7,152 | 1,785 | -75% | 1 | 1 | 0% | 1,314 | 551 | -58% | 0 | 0 | — |
case-15 | fail→pass | 2,255 | 1,926 | -15% | 1 | 1 | 0% | 261 | 557 | +113% | 0 | 0 | — |
case-16 | fail→pass | 2,376 | 1,409 | -41% | 1 | 1 | 0% | 362 | 519 | +43% | 0 | 0 | — |
case-17 | pass→pass | 7,364 | 2,323 | -68% | 1 | 1 | 0% | 1,214 | 652 | -46% | 0 | 0 | — |
case-18 | pass→pass | 5,843 | 2,102 | -64% | 1 | 1 | 0% | 1,030 | 578 | -44% | 0 | 0 | — |
case-19 | pass→pass | 4,534 | 1,676 | -63% | 1 | 1 | 0% | 747 | 550 | -26% | 0 | 0 | — |
case-20 | pass→pass | 15,190 | 13,219 | -13% | 1 | 1 | 0% | 2,914 | 2,857 | -2% | 0 | 0 | — |
case-21 | pass→pass | 12,576 | 8,316 | -34% | 1 | 1 | 0% | 2,391 | 1,772 | -26% | 0 | 0 | — |
case-22 | pass→pass | 7,265 | 3,604 | -50% | 1 | 1 | 0% | 1,402 | 900 | -36% | 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.
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.