jiushiwon/wg-skills
Developer toolsThis page was translated automatically. Technical details may differ from the original. View the original
My Good AI SKILLS
Overview
Any agent (Claude / Codex / Workbuddy / OpenCode / VS Code / Cursor / Cline / Kimi, etc.) can load and use the skills of this repository. Programming in the AI era is not about "replacing programmers," but about extracting the most important parts of traditional engineering and distilling them into skills that can be called by AI, allowing humans to focus on business decisions and architectural judgments. One-sentence selection + one-click generation + business module stacking, covering four major languages: Java, Python, Go, Node.js. Covers uni-app / Vue / general frontend, with component libraries, request layers, and style specifications all fully integrated. : Covers a large number of middleware deployments (Redis / Kafka / MySQL / PostgreSQL / MongoDB / Go / Java / Python / Node.js). Supports vibeCoding/super-deploy-skills/, covering installation and orchestration of Redis / Kafka / MySQL / Postgres / Mongodb / Go / Java / Python / Nodejs and others. : This repository is not "AI helped me write a project," but rather "includes complete front-end and back-end and database data, including architecture." — All project-level logic (development principles, creating/modifying skill processes, commit specifications, output requirements, parent-child nested structure descriptions) are all here.
Overview (machine translated)
158 public Skills MIT protocols for agents such as Claude Code / Cursor / Codex. is not a collection of prompts, but a foundational capability system that enables AI to complete “project skeleton→ modules → front-end and back-end contracts → deployment → iterations” according to unified engineering standards.
About itlifetime.com
itlifetime.com is the public portal for “Koala AI Development”: here it continuously displays real projects, page effects, usage paths for wg-skills, and a series of articles.
| The problem you are solving | Where to start |
|---|---|
| Want to create a run-worthy project | See Project Cases and Services |
| Want to understand this public system from scratch | See the wg-skills skill catalog and ONBOARDING.md |
| If you want to build your project experience into a Skill | First, read the “Who is suitable for this” below, then start by reviewing a real project |
I don’t sell “universal one-click generation” here. What I’m sharing is an engineering habit repeatedly validated over ten years of full-stack development: first establish common rules, then let AI efficiently complete the specific tasks.
What is this?
A self-containing skill repository. Each subdirectory is a skill (SKILL.md + README.md + references/), and they are not isolated: frontend pages and themes, backend skeleton and modules, database and interface contracts, deployment and acceptance can be combined as needed by the project.
It accumulates common elements that repeatedly appear in full-stack projects: directory structure, response format, error codes, JWT, pagination, Design Token, request layer, environment and deployment. Specific industry business, brand vision, and product trade-offs still have to be decided by the project leader.
One-sentence positioning
** is not about letting AI do projects for you, but rather letting AI do projects based on your engineering experience. **
extracting the repetitive, standardized, and verifiable parts from traditional full-stack development into AI-usable skills; while humans are still responsible for business boundaries, product judgment, and final acceptance.
Differences from “Traditional AI One-Click Project Creation”
| Dimension | Traditional AI One-Click Project Execution | wg-skills |
|---|---|---|
| Output | One-time code snippets | Complete, runnable, and evolving projects |
| Front-End and Backend | Generate only one end | Complete front-end and back-end + database, connected via api-contract.md contract |
| Documentation | Usually none | Forced delivery api-contract.md + docs/project-guide.md |
| Frontend Architecture | Single Framework | 4 Systems: Vue / UniApp / React / HTML, sharing Design Token and request layer |
| Standard | AI Free Play | Skill Mandatory Constraints (Generate and Comply) |
This page was translated automatically. Technical details may differ from the original.
Recommended Tools
Try a different keyword or remove a filter.
Install
npx skillfish add jiushiwon/wg-skills