Skip to content

wigtn/wigtn-plugins-with-claude-code

Folders and files

NameName
Last commit message
Last commit date

Latest commit

Β 

History

87 Commits
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 

Repository files navigation

English | ν•œκ΅­μ–΄

WIGTN Coding

From Idea to Deploy, Zero Friction

Version Agents Skills Styles

GitHub Stars License Contributors Last Commit


What is WIGTN Coding?

WIGTN Coding is a single, unified Claude Code plugin that takes you from a rough idea to a deployed product with zero friction. One plugin β€” no prefixes needed.

"I want to build a SaaS dashboard with user auth"
  β†’ /prd        (requirements)
  β†’ digging     (analysis)
  β†’ /implement  (parallel build)
  β†’ /auto-commit (quality gate + commit)

12 agents, 3 commands, 3 skills, 17 design styles β€” all working together with team-based parallel execution for 3-5x speedup.


At a Glance

What Count Highlights
Agents 12 Parallel coordinators, architecture decisions, specialized developers
Commands 3 /prd, /implement, /auto-commit
Skills 3 Code review levels, design system reference, team memory protocol
Design Styles 17 Editorial, Brutalist, Glassmorphism, Liquid Glass, Claymorphism, and more
Hooks 4 Dangerous command blocking, formatting reminders, pattern compliance

Installation

# Step 1 β€” Add marketplace source (first time only)
/plugin marketplace add wigtn/wigtn-plugins-with-claude-code

# Step 2 β€” Install plugin
/install wigtn-coding
Manual install (alternative)
git clone https://github.com/wigtn/wigtn-plugins-with-claude-code.git ~/.claude-plugins/wigtn
mkdir -p ~/.claude/plugins
ln -s ~/.claude-plugins/wigtn/plugins/wigtn-coding ~/.claude/plugins/

# Update
git -C ~/.claude-plugins/wigtn pull

The Pipeline

The core workflow is a 4-step pipeline that goes from idea to committed code:

β”Œβ”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”
β”‚                                                                 β”‚
β”‚   /prd "user authentication with OAuth"                         β”‚
β”‚   β”œβ”€β”€ PRD.md (structured requirements)                          β”‚
β”‚   └── PLAN_{feature}.md (phased task plan)                      β”‚
β”‚                         ↓                                       β”‚
β”‚   digging (4-agent parallel analysis)                           β”‚
β”‚   β”œβ”€β”€ Completeness β€” missing requirements?                      β”‚
β”‚   β”œβ”€β”€ Feasibility β€” can we actually build this?                 β”‚
β”‚   β”œβ”€β”€ Security β€” any vulnerabilities?                           β”‚
β”‚   └── Consistency β€” do requirements contradict?                 β”‚
β”‚                         ↓                                       β”‚
β”‚   /implement --parallel                                         β”‚
β”‚   β”œβ”€β”€ DESIGN Phase (3 agents parallel)                          β”‚
β”‚   β”‚   β”œβ”€β”€ PRD search + quality gate                             β”‚
β”‚   β”‚   β”œβ”€β”€ Architecture decision (MSA vs Monolith)               β”‚
β”‚   β”‚   └── Project analysis + gap analysis                       β”‚
β”‚   β”œβ”€β”€ User approval checkpoint                                  β”‚
β”‚   └── BUILD Phase (team-based parallel)                         β”‚
β”‚       β”œβ”€β”€ Backend team  β†’ backend-architect agent               β”‚
β”‚       β”œβ”€β”€ Frontend team β†’ frontend-developer agent              β”‚
β”‚       β”œβ”€β”€ AI team       β†’ ai-agent (when needed)                β”‚
β”‚       └── Ops team      β†’ devops setup (when needed)            β”‚
β”‚                         ↓                                       β”‚
β”‚   /auto-commit                                                  β”‚
β”‚   β”œβ”€β”€ 3-agent parallel code review                              β”‚
β”‚   β”œβ”€β”€ Quality Gate (score 80+ = auto-commit)                    β”‚
β”‚   β”œβ”€β”€ Security zero-tolerance check                             β”‚
β”‚   └── Commit + push                                             β”‚
β”‚                                                                 β”‚
β””β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”˜

Quality Gate

Score Action
80+ Auto-commit
60-79 Auto-fix then retry
< 60 Block commit
Security Critical Force FAIL (capped at 59)

Parallel Speedup

Stage Sequential Parallel Speedup
digging 4 categories serial 4 agents parallel 4x
DESIGN 4 steps serial 3 agents parallel 3x
BUILD tasks serial team-based parallel 2-3x
Review single reviewer 3 agents parallel 3x
Full Pipeline ~20 min ~6 min ~3x

Commands (3)

Command Description
/prd <feature> Generate PRD + phased task plan from a feature idea
/implement <feature> Design + build with automatic parallel mode detection
/implement --parallel Force parallel team-based build
/auto-commit Parallel quality review + safety gate + auto-commit

Domain-specific capabilities (backend, frontend, mobile, AI, DevOps) are handled by specialized agents dispatched automatically through /implement's team-build-coordinator.


Agents (12)

Agent Role
architecture-decision Analyzes PRD to decide MSA vs Monolithic vs Modular Monolith
code-formatter Multi-language formatting and linting automation
code-reviewer Code review with 100-point quality scoring (readability, maintainability, performance, testability, best practices)
prd-reviewer PRD analysis β€” finds weaknesses, gaps, and risks across 4 categories (completeness, feasibility, security, consistency)
team-build-coordinator Team-based parallel build: Backend + Frontend + AI + Ops
parallel-review-coordinator 3-agent parallel code review with score merge
parallel-digging-coordinator 5-phase parallel PRD analysis pipeline with quality gate
frontend-developer React 19 / Next.js 16+ component and page generation with 17 design styles
design-discovery VS (Verbalized Sampling) style recommendation for Web and Mobile
backend-architect Backend patterns, API design, database schema, architecture decisions
mobile-developer React Native / Expo component, screen, and native module generation
ai-agent STT (WhisperX) and LLM (OpenAI, Anthropic) integration patterns

Skills (3)

Skill Description
code-review-levels Reference documents for deep code review (Level 3: call chains, edge cases, concurrency, security) and architecture review (Level 4: SOLID principles, dependency analysis, layer violations, scalability)
design-system-reference 17 style guides with common patterns (animations, colors, spacing). Works with design-discovery agent for context gathering and VS-based style recommendations. Includes anti-patterns and implementation checklists
team-memory-protocol Shared context management for team-based parallel builds. SHARED_CONTEXT file management, TaskCreate integration, and cross-agent memory coordination

Design Styles (17)

The design-system-reference skill includes 17 professionally crafted style guides. Each style guide covers philosophy, typography, layout, color, components, motion, and anti-patterns.

Style Vibe
Editorial Magazine-inspired layouts with strong serif typography
Brutalist Raw, bold, unconventional β€” breaks all the rules
Glassmorphism Frosted glass effects with blur and transparency
Swiss Minimal Clean grid-based design, typography-focused
Neomorphism Soft UI with subtle inset/outset shadows
Bento Grid Modern card-based grid layouts (Apple-inspired)
Dark Mode First Designed for dark interfaces from the ground up
Minimal Corporate Clean, professional business aesthetic
Retro Pixel CRT effects, monospace fonts, terminal nostalgia
Organic Shapes Blob shapes, natural curves, earthy tones
Maximalist Bold typography, intense colors, layered complexity
3D Immersive CSS 3D transforms, parallax, depth effects
Liquid Glass Fluid, translucent glass effects with dynamic reflections
Claymorphism Soft 3D clay-like elements with pastel tones
Minimalism Extreme simplicity, whitespace-driven, essential-only
Neobrutalism Modern brutalism with colorful accents and bold borders
Skeuomorphism Realistic textures and physical-world metaphors

The design-discovery agent uses VS (Verbalized Sampling) technique to recommend the best style for your project context.


Hooks (Safety & Quality)

Hook Trigger What It Does
Dangerous Command Blocker Bash PreToolUse Blocks rm -rf /, git push --force, DROP TABLE, etc.
Pipeline Completion Stop event Reminds you to review changes before pushing
Frontend Formatting Write|Edit PostToolUse Reminds to run prettier/eslint on .tsx, .jsx, .css files
Backend Pattern Compliance Write|Edit PostToolUse Reminds to verify error handling, input validation, logging on .ts, .py, .go files

Scenarios

Scenario 1: Full-Stack SaaS App from Scratch

"I want to build a project management tool with team collaboration"

# Step 1 β€” Define requirements
/prd project management tool with kanban boards and team collaboration

# Step 2 β€” Review and refine the plan (runs digging automatically)
# 4-agent parallel analysis catches gaps: "Missing: real-time sync, role permissions"

# Step 3 β€” Build everything in parallel
/implement --parallel project-management
# Backend team: API endpoints, Prisma schema, auth middleware
# Frontend team: Kanban board, team views, dashboard
# Ops team: Dockerfile, GitHub Actions CI/CD

# Step 4 β€” Quality check and commit
/auto-commit
# 3 reviewers score: 87/100 β†’ Auto-commit

Scenario 2: Mobile App Development

"Build a fitness tracking app with React Native"

/prd fitness tracker with workout logging, progress charts, and Apple Health sync

/implement fitness-tracker
# Architecture: Expo Router + Zustand + MMKV + React Query
# Mobile-specific: biometric auth, haptic feedback, offline sync
# Generates: screens, components, navigation, native module integration

Scenario 3: Design-Driven Landing Page

"Create a landing page for my AI startup"

/prd AI startup landing page with modern design

/implement ai-landing
# design-discovery agent activates β†’ asks about brand personality
# Recommends: Glassmorphism (modern, trust) or Liquid Glass (dynamic, cutting-edge)
# Frontend team generates: Hero, Features, Pricing, CTA sections with chosen style

/auto-commit

Scenario 4: Backend API with AI Features

"Build a transcription service with voice commands and LLM processing"

/prd transcription service with WhisperX STT and LLM-powered summarization

/implement --parallel transcription-service
# Backend team β†’ backend-architect: API endpoints, database schema, auth
# AI team β†’ ai-agent: WhisperX integration, OpenAI/Anthropic API patterns
# Frontend team β†’ frontend-developer: Upload UI, transcription viewer, dashboard
# Ops team: Docker setup, CI/CD pipeline

/auto-commit
# Quality gate: 92/100 β†’ Auto-commit

Scenario 5: DevOps Pipeline Setup

"Set up Docker and CI/CD for my monorepo"

/prd Docker multi-stage build and GitHub Actions CI/CD for pnpm monorepo

/implement devops-pipeline
# team-build-coordinator dispatches ops team tasks
# Generates: Dockerfile, docker-compose.yml, .github/workflows/ci.yml
# Includes: caching, test stage, deployment to Vercel/Railway

/auto-commit

Plugin Structure

plugins/wigtn-coding/
β”œβ”€β”€ .claude-plugin/
β”‚   └── plugin.json           # Plugin metadata (12 agents, 3 commands, 3 skills)
β”œβ”€β”€ agents/                   # 12 agent definitions
β”‚   β”œβ”€β”€ architecture-decision.md
β”‚   β”œβ”€β”€ code-formatter.md
β”‚   β”œβ”€β”€ code-reviewer.md
β”‚   β”œβ”€β”€ prd-reviewer.md
β”‚   β”œβ”€β”€ team-build-coordinator.md
β”‚   β”œβ”€β”€ parallel-review-coordinator.md
β”‚   β”œβ”€β”€ parallel-digging-coordinator.md
β”‚   β”œβ”€β”€ frontend-developer.md
β”‚   β”œβ”€β”€ design-discovery.md
β”‚   β”œβ”€β”€ backend-architect.md
β”‚   β”œβ”€β”€ mobile-developer.md
β”‚   └── ai-agent.md
β”œβ”€β”€ commands/                 # 3 user-invocable commands
β”‚   β”œβ”€β”€ prd.md
β”‚   β”œβ”€β”€ implement.md
β”‚   └── auto-commit.md
β”œβ”€β”€ skills/                   # 3 skills with reference files
β”‚   β”œβ”€β”€ code-review-levels/   # Deep review (Level 3) + architecture review (Level 4)
β”‚   β”‚   β”œβ”€β”€ SKILL.md
β”‚   β”‚   β”œβ”€β”€ deep-review.md
β”‚   β”‚   └── architecture-review.md
β”‚   β”œβ”€β”€ design-system-reference/  # 17 style guides + common patterns
β”‚   β”‚   β”œβ”€β”€ SKILL.md
β”‚   β”‚   β”œβ”€β”€ README.md
β”‚   β”‚   β”œβ”€β”€ common/
β”‚   β”‚   β”‚   β”œβ”€β”€ animations.md
β”‚   β”‚   β”‚   β”œβ”€β”€ colors.md
β”‚   β”‚   β”‚   └── spacing.md
β”‚   β”‚   └── styles/           # 17 design style guides
β”‚   β”‚       β”œβ”€β”€ editorial.md
β”‚   β”‚       β”œβ”€β”€ brutalist.md
β”‚   β”‚       β”œβ”€β”€ glassmorphism.md
β”‚   β”‚       β”œβ”€β”€ swiss-minimal.md
β”‚   β”‚       β”œβ”€β”€ neomorphism.md
β”‚   β”‚       β”œβ”€β”€ bento-grid.md
β”‚   β”‚       β”œβ”€β”€ dark-mode-first.md
β”‚   β”‚       β”œβ”€β”€ minimal-corporate.md
β”‚   β”‚       β”œβ”€β”€ retro-pixel.md
β”‚   β”‚       β”œβ”€β”€ organic-shapes.md
β”‚   β”‚       β”œβ”€β”€ maximalist.md
β”‚   β”‚       β”œβ”€β”€ 3d-immersive.md
β”‚   β”‚       β”œβ”€β”€ liquid-glass.md
β”‚   β”‚       β”œβ”€β”€ claymorphism.md
β”‚   β”‚       β”œβ”€β”€ minimalism.md
β”‚   β”‚       β”œβ”€β”€ neobrutalism.md
β”‚   β”‚       └── skeuomorphism.md
β”‚   └── team-memory-protocol/ # Cross-agent shared context
β”‚       └── SKILL.md
└── hooks/
    └── hooks.json            # 4 hooks (safety + quality)

Tech Stack Coverage

Domain Technologies
Frontend React 19, Next.js 16+, Tailwind CSS, Radix UI, React Hook Form, Zod
Backend NestJS, Express, Fastify, FastAPI, Prisma, TypeORM, Drizzle
Mobile React Native 0.73+, Expo SDK 52+, Expo Router, React Navigation
Database PostgreSQL, MySQL, MongoDB, SQLite
State Zustand, Jotai, Redux Toolkit, React Query, MMKV
Testing Jest, RTL, RNTL, Playwright, Detox, Maestro, MSW
DevOps Docker, Kubernetes, GitHub Actions, Vercel, Railway
AI WhisperX (STT), OpenAI GPT, Anthropic Claude
Design 17 style systems, VS-based style discovery, HIG, Material Design 3

Contributing

  1. Fork the repository
  2. Create your feature branch (git checkout -b feature/amazing-skill)
  3. Commit your changes (git commit -m 'feat: Add amazing skill')
  4. Push to the branch (git push origin feature/amazing-skill)
  5. Open a Pull Request

License

This project is licensed under the MIT License β€” see the LICENSE file for details.


Made with Claude Code by WIGTN Crew

About

Claude Code Plugins for AI Native-based Junior Developer

Topics

Resources

License

Stars

Watchers

Forks

Releases

No releases published

Packages

 
 
 

Contributors