Node.js (Backend) Skill
Goal
Strict-typed, layered, testable Node services. No any, no callback soup,
no magic strings. A junior should be able to trace a request end-to-end.
Quick reference
| Concept |
Best practice |
| TypeScript |
Use strict mode, ESM imports ("type": "module"), correct path aliases |
| Frameworks |
Fastify or NestJS with ChangeDetectionStrategy / proper DI |
| Performance |
Do not block the event loop, use worker threads for CPU-heavy tasks |
| Key commands |
npm run dev, npm run build, npm test, npx tsc --noEmit |
Full guidance
Extended how-to, patterns, anti-patterns, and checklists: SKILL.deep.md
Source: PetrovC/ai-agent-kit — distributed by TomeVault.
1---2name: petrovc-ai-agent-kit-node3description: Node.js (Backend) Skill4---56# Node.js (Backend) Skill78## Goal9Strict-typed, layered, testable Node services. No `any`, no callback soup,10no magic strings. A junior should be able to trace a request end-to-end.1112## Quick reference1314| Concept | Best practice |15|---|---|16| TypeScript | Use strict mode, ESM imports (`"type": "module"`), correct path aliases |17| Frameworks | Fastify or NestJS with ChangeDetectionStrategy / proper DI |18| Performance | Do not block the event loop, use worker threads for CPU-heavy tasks |19| Key commands | `npm run dev`, `npm run build`, `npm test`, `npx tsc --noEmit` |2021## Full guidance22Extended how-to, patterns, anti-patterns, and checklists: [`SKILL.deep.md`](SKILL.deep.md)2324---25> Source: [PetrovC/ai-agent-kit](https://github.com/PetrovC/ai-agent-kit) — distributed by [TomeVault](https://tomevault.io).26<!-- tomevault:4.0:skill_md:2026-06-15 -->
tomevault-io