node-programming

name	node-programming
description	Node.js and JavaScript/TypeScript development guidance for creating, modifying, reviewing, testing, packaging, and debugging backend, CLI, tooling, and full-stack code. Use when tasks involve package.json, npm, pnpm, yarn, bun, .js, .mjs, .cjs, .ts, tsconfig, Express, NestJS, Next.js API routes, Vitest, Jest, ESLint, or Node dependency management.

Node Programming

Goal

Make Node.js changes that respect the project's runtime, module system, package manager, and TypeScript strictness.

Discovery

• Identify the package manager from lockfiles: package-lock.json, pnpm-lock.yaml, yarn.lock, or bun.lockb.
• Read package.json scripts, tsconfig*.json, lint config, test config, and framework conventions before editing.
• Preserve the existing module style: ESM, CommonJS, or mixed boundaries.
• Check whether the code targets Node only, browser bundles, serverless, or a full-stack framework.

Implementation

• Prefer typed, explicit data flow in TypeScript; avoid weakening types with any unless the surrounding code already accepts it.
• Keep async code awaitable and handle rejected promises.
• Preserve environment variable conventions and validation patterns.
• Avoid switching package managers or adding dependencies unless the need is clear and the existing stack supports it.
• For HTTP APIs, keep route, validation, service, and test changes aligned.
• For CLIs, preserve exit codes, stdout/stderr conventions, and non-interactive behavior.

Verification

• Use the project's scripts first: npm test, pnpm test, yarn test, bun test, npm run lint, npm run typecheck, or npm run build.
• Prefer targeted tests before full suites when the project is large.
• If dependencies are missing, run syntax or TypeScript checks that are available and report the gap clearly.

Review Focus

• ESM/CJS mismatches, unhandled promises, accidental browser-only or Node-only imports, broken build scripts, unsafe environment access, dependency churn, and generated lockfile noise.