Nodejs Project Arch — Data skill for Claude Code

nodejs-project-arch

**Node.js project architecture standards for AI-assisted development.**

An [OpenClaw](https://openclaw.ai) / AI agent skill that enforces modular file structures, keeping every file small enough for AI to read and edit without blowing the context window.

✨ Why?

When AI agents work with large single-file codebases, things go wrong fast:

Scenario	Tokens per read	Context usage (200K)
3000-line server.js	~40K	20%
200-line module	~2.7K	1.3%

**Result:** Splitting files means **10-15 productive rounds** before context compression, vs 3-5 with monolithic files. That's **70-93% token savings** per file read.

📐 Core Rules

• Single file ≤ 400 lines / index.html ≤ 200 lines / server.js entry ≤ 100 lines
• All tunable values in config.json — loaded at runtime, editable via admin dashboard
• Backend: routes/ by domain, services/ for shared logic, db.js for database init
• Frontend: HTML skeleton only, JS/CSS in separate files under public/
• Every project gets admin.html + routes/admin.js for config hot-reload + stats

🎮 Supported Project Types

Type	Signals	Reference
H5 Game	Canvas, Phaser, Matter.js, game loop, sprites	references/game.md
Data Tool	Crawler, scraper, scheduler, data sync, analytics	references/tool.md
Content/Utility	Generator, library, publisher, file processing	references/tool.md
Dashboard/Monitor	Charts, real-time, alerts, metrics	references/tool.md
API Service	REST endpoints, middleware, microservice	references/tool.md
SDK/Library	Shared module, build step, multi-consumer	references/sdk.md

📁 Typical Structure

project-name/
  server.js          ← Entry point,