ateschh-kit 1.0.0

package/CLAUDE.md

@@ -0,0 +1,154 @@
+# ATESCHH KIT — AI Development System
+
+## Identity
+
+You are an AI software development assistant.
+Make technical decisions. Explain your choices clearly.
+You build — you don't tell the user to build.
+
+Rules: `.claude/rules/` (auto-loaded)
+
+---
+
+## Session Start Protocol
+
+On every new session, FIRST:
+
+1. Read `.state/ACTIVE-PROJECT.md`
+2. If active project exists → read `projects/{name}/STATE.md`
+3. Give user a 3-line summary:
+   - "📁 Active project: {name}"
+   - "✅ Last completed: {task}"
+   - "➡️ Next up: {task}"
+
+If no active project:
+> "No active project. Use `/new-project` to start or `/resume` to return to an existing one."
+
+---
+
+## Slash Commands
+
+| Command | What it does | Workflow |
+|---------|-------------|---------|
+| `/new-project` | Start a new project | workflows/new-project.md |
+| `/resume` | Continue where you left off | workflows/resume.md |
+| `/brainstorm` | Idea analysis + market research | workflows/brainstorm.md |
+| `/requirements` | Define and lock tech stack | workflows/requirements.md |
+| `/design` | Pages + UI design | workflows/design.md |
+| `/build` | Write page/module code | workflows/build.md |
+| `/test` | Test + fix bugs | workflows/test.md |
+| `/deploy` | Deploy to production | workflows/deploy.md |
+| `/status` | Progress report | workflows/status.md |
+| `/save` | Save context (cross-platform) | workflows/save.md |
+| `/finish` | Complete and archive project | workflows/finish.md |
+| `/next` | Auto-detect and run next step | workflows/next.md |
+| `/quick` | Ad-hoc task without full pipeline | workflows/quick.md |
+| `/map-codebase` | Analyze existing codebase | workflows/map-codebase.md |
+| `/settings` | View/edit configuration | workflows/settings.md |
+
+When a command is run, read the corresponding workflow file and follow its steps.
+
+---
+
+## Project Phases
+
+```
+1. /brainstorm      → Idea analysis + market research
+2. /requirements    → Tech stack locked
+3. /design          → Pages + design locked
+4. /build           → Build page by page (iterative)
+5. /test            → Test + fix bugs
+6. /deploy          → Go live
+```
+
+---
+
+## Golden Rules (Immutable)
+
+1. **ONE PROJECT**: Only one active project at a time
+2. **PLAN FIRST**: Approve PLAN.md before writing code
+3. **REQUIREMENTS LOCK**: Libraries in REQUIREMENTS.md cannot be changed
+4. **SMALL STEPS**: Each task is max 1 hour of work
+5. **SAVE**: Update STATE.md after every task
+6. **SHOW**: Show the result of every action to the user
+7. **BACKLOG**: New ideas go to BACKLOG.md — not now
+
+---
+
+## Agent System
+
+Use specialist agents for complex tasks.
+Agent definitions: `agents/` folder.
+
+| Agent | Role | Triggered by |
+|-------|------|-------------|
+| idea-analyst | Analyzes idea with Socratic method | /brainstorm |
+| market-researcher | Competitor and market research | /brainstorm |
+| requirements-expert | Selects and locks tech stack | /requirements |
+| architect | Page structure and features | /design |
+| designer | Colors, fonts, UI system | /design |
+| coder | Implements code | /build |
+| tester | Writes and runs tests | /test |
+| deployer | Deploys to production | /deploy |
+| debugger | Finds and fixes bugs | /test, /build |
+
+**Auto-detection**: The system automatically selects the best specialist based on your request. You don't need to mention agent names explicitly.
+
+---
+
+## Skill System
+
+Reusable atomic tasks: `skills/`
+
+| Skill | Role |
+|-------|------|
+| idea-analysis | 5-question idea analysis |
+| market-research | 3-5 competitor research |
+| requirements-lock | Write stack decisions to REQUIREMENTS.md |
+| architecture-design | Create STRUCTURE.md + DESIGN.md |
+| write-code | Implement a PLAN.md task |
+| run-tests | L1-L4 validation |
+| fix-bugs | Bug detection + fix |
+| publish | Deploy via MCP |
+| context-management | Save/load/sync session |
+
+---
+
+## Quality Levels
+
+| Level | Check | Required |
+|-------|-------|---------|
+| L1 | Syntax | Always |
+| L2 | Functionality | Always |
+| L3 | Integration | At /test |
+| L4 | Quality | Before /deploy |
+
+**Do not proceed to the next task without passing L2.**
+
+---
+
+## Cross-Platform Usage (Claude Code ↔ Antigravity)
+
+When switching platforms:
+1. On current platform: `/save`
+2. On new platform: open the ATESCHH KIT directory
+3. MEMORY.md is auto-loaded
+4. Type `/resume` → continue where you left off
+
+---
+
+## File System
+
+```
+ATESCHH-KIT/
+├── CLAUDE.md              ← This file
+├── .claude/rules/         ← Auto-loaded rules
+├── agents/                ← Agent definitions
+├── skills/                ← Skill definitions
+├── workflows/             ← Slash command logic
+├── context-agent/         ← Context management
+├── .state/                ← System state (gitignored)
+├── templates/             ← Project templates
+├── projects/              ← Active projects (gitignored)
+└── archive/               ← Completed projects (gitignored)
+```

package/CONTRIBUTING.md

@@ -0,0 +1,53 @@
+# Contributing to ateschh-kit
+
+Thank you for your interest in contributing!
+
+## Ways to Contribute
+
+- **Bug reports**: Open an issue describing the behavior and expected behavior
+- **New workflows**: Add a new slash command in `workflows/`
+- **New agents**: Add a specialist agent in `agents/`
+- **New skills**: Add a reusable skill in `skills/`
+- **Documentation**: Improve or translate README files
+- **Stack support**: Add deployment playbooks for new platforms
+
+## Guidelines
+
+### Adding a Workflow
+
+1. Copy `workflows/_TEMPLATE.md`
+2. Fill in all sections
+3. Add the command to `CLAUDE.md` command table
+4. Test it with a real project
+
+### Adding an Agent
+
+1. Copy `agents/_TEMPLATE.md`
+2. Define: role, process, output format, rules
+3. Reference it from the appropriate workflow
+4. Document which skill(s) it uses
+
+### Adding a Skill
+
+1. Create `skills/{name}/SKILL.md`
+2. Keep it atomic — one skill, one responsibility
+3. Document: purpose, steps, output template
+
+## Principles
+
+- **Stay thin**: Orchestrators coordinate. Agents and skills do the work.
+- **Explicit over implicit**: Every rule, every constraint, every output format should be written down.
+- **English only**: All system files are in English. Translations live in README.{lang}.md files.
+- **No personal references**: All content must work for any user.
+
+## Pull Request Process
+
+1. Fork the repository
+2. Create a branch: `git checkout -b feature/your-feature`
+3. Make your changes
+4. Test with at least one real project
+5. Submit a PR with a clear description
+
+## License
+
+By contributing, you agree that your contributions will be licensed under the MIT License.

package/LICENSE

@@ -0,0 +1,21 @@
+MIT License
+
+Copyright (c) 2026 ateschh-kit contributors
+
+Permission is hereby granted, free of charge, to any person obtaining a copy
+of this software and associated documentation files (the "Software"), to deal
+in the Software without restriction, including without limitation the rights
+to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
+copies of the Software, and to permit persons to whom the Software is
+furnished to do so, subject to the following conditions:
+
+The above copyright notice and this permission notice shall be included in all
+copies or substantial portions of the Software.
+
+THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
+AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
+LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
+OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
+SOFTWARE.

package/README.md

@@ -0,0 +1,145 @@
+# ateschh-kit
+
+> A structured AI development system for Claude Code and Antigravity.  
+> Goes from idea to deployment without context rot or project abandonment.
+
+---
+
+## Install
+
+```bash
+npx ateschh-kit
+```
+
+That's it. The system is now installed in your current directory.
+
+---
+
+## What This Is
+
+A structured workflow system that guides AI agents through the full software development lifecycle:
+
+```
+/brainstorm → /requirements → /design → /build → /test → /deploy
+```
+
+Each phase is gated. You can't accidentally skip from brainstorm straight to coding.
+
+## Why It Works
+
+| Problem | Solution |
+|---------|---------|
+| AI forgets context between sessions | `/save` + `/resume` — works across Claude Code and Antigravity |
+| Mid-project "let's switch frameworks" | REQUIREMENTS.md is locked after approval |
+| Scope creep | New ideas go to BACKLOG.md — not now |
+| Project abandonment | STATE.md always knows what's next |
+| "Just fix this one thing" rabbit holes | `/quick` for ad-hoc tasks, `/next` for auto-pilot |
+
+---
+
+## Commands
+
+| Command | What it does |
+|---------|-------------|
+| `/new-project` | Start a new project |
+| `/brainstorm` | Analyze your idea + research the market |
+| `/requirements` | Select and lock the tech stack |
+| `/design` | Define pages, features, and visual system |
+| `/build` | Implement one task from the plan |
+| `/test` | Run L1–L4 quality checks |
+| `/deploy` | Deploy to production |
+| `/status` | See where you are |
+| `/save` | Save context (cross-platform) |
+| `/resume` | Continue from last session |
+| `/next` | Auto-detect and run the right next step |
+| `/quick` | Ad-hoc task without the full pipeline |
+| `/map-codebase` | Analyze an existing codebase |
+| `/settings` | View/edit configuration |
+
+---
+
+## How to Use
+
+**Starting a new project:**
+```
+/new-project
+/brainstorm  ← describe your idea
+/requirements
+/design
+/build  ← repeat until done
+/test
+/deploy
+```
+
+**Returning to a project:**
+```
+/resume
+# or
+/next  ← automatically detects where you left off
+```
+
+**Working on an existing codebase:**
+```
+/map-codebase
+```
+
+---
+
+## What's Inside
+
+```
+ateschh-kit/
+├── CLAUDE.md              ← Main orchestration file
+├── .claude/rules/         ← 7 auto-loaded behavioral rules
+├── agents/                ← 9 specialist agents
+├── skills/                ← 9 reusable atomic skills
+├── workflows/             ← 15 slash-command workflows
+├── templates/             ← Project file templates
+└── context-agent/         ← Context management system
+```
+
+---
+
+## Supported Platforms
+
+| Platform | Status |
+|----------|--------|
+| Claude Code | ✅ Full support |
+| Antigravity | ✅ Full support |
+| Cursor | ✅ Works (agents folder) |
+| Windsurf | ✅ Works (agents folder) |
+
+---
+
+## Supported Stacks
+
+| Type | Default Stack |
+|------|--------------|
+| Web App | Next.js + Supabase + Vercel |
+| Mobile | Expo + Supabase |
+| Browser Extension | Plasmo + TypeScript |
+| Backend API | Hono + Cloudflare Workers |
+| Desktop | Electron / Tauri |
+
+---
+
+## Cross-Platform Context
+
+Work on the same project from Claude Code and Antigravity interchangeably:
+
+```
+Platform A → /save
+Platform B → /resume   ← picks up exactly where you left off
+```
+
+---
+
+## License
+
+MIT — use freely, modify freely, no attribution required.
+
+---
+
+## Contributing
+
+See [CONTRIBUTING.md](CONTRIBUTING.md).

package/README.tr.md

@@ -0,0 +1,145 @@
+# ateschh-kit
+
+> Claude Code ve Antigravity için yapılandırılmış bir AI geliştirme sistemi.  
+> Fikirden dağıtıma kadar, bağlam kaybı ve proje terk etme olmadan çalışır.
+
+---
+
+## Kurulum
+
+```bash
+npx ateschh-kit
+```
+
+Hepsi bu. Sistem bulunduğunuz dizine kurulur.
+
+---
+
+## Bu Nedir?
+
+AI ajanlarını yazılım geliştirme sürecinin tamamında yönlendiren yapılandırılmış bir iş akışı sistemi:
+
+```
+/brainstorm → /requirements → /design → /build → /test → /deploy
+```
+
+Her aşama geçiş gerektiriyor. Brainstorm'dan direkt kodlamaya geçemezsiniz.
+
+## Neden İşe Yarıyor?
+
+| Sorun | Çözüm |
+|-------|-------|
+| AI oturumlar arasında bağlamı unutuyor | `/save` + `/resume` — Claude Code ve Antigravity arasında çalışır |
+| Proje ortasında "gelin framework değiştirelim" | REQUIREMENTS.md onay sonrası kilitlenir |
+| Kapsam kayması | Yeni fikirler BACKLOG.md'ye gider — şimdi değil |
+| Proje terk etme | STATE.md her zaman sıradakini bilir |
+| "Şunu hızlıca düzelt" tuzakları | `/quick` tek seferlik görevler için, `/next` otomatik pilot için |
+
+---
+
+## Komutlar
+
+| Komut | Ne Yapar |
+|-------|---------|
+| `/new-project` | Yeni proje başlat |
+| `/brainstorm` | Fikir analizi + pazar araştırması |
+| `/requirements` | Teknoloji yığınını seç ve kilitle |
+| `/design` | Sayfalar, özellikler ve versal sistem |
+| `/build` | Plandan bir görevi uygula |
+| `/test` | L1–L4 kalite kontrolleri |
+| `/deploy` | Üretim ortamına dağıt |
+| `/status` | Nerede olduğunu gör |
+| `/save` | Bağlamı kaydet (çapraz platform) |
+| `/resume` | Son oturumdan devam et |
+| `/next` | Doğru sonraki adımı otomatik tespit et ve çalıştır |
+| `/quick` | Tam pipeline olmadan tek seferlik görev |
+| `/map-codebase` | Mevcut projeyi analiz et |
+| `/settings` | Yapılandırmayı görüntüle/düzenle |
+
+---
+
+## Nasıl Kullanılır
+
+**Yeni proje başlatmak:**
+```
+/new-project
+/brainstorm  ← fikrinizi anlatın
+/requirements
+/design
+/build  ← tamamlanana kadar tekrarlayın
+/test
+/deploy
+```
+
+**Bir projeye dönmek:**
+```
+/resume
+# veya
+/next  ← kaldığınız yeri otomatik tespit eder
+```
+
+**Mevcut bir proje üzerinde çalışmak:**
+```
+/map-codebase
+```
+
+---
+
+## İçerik
+
+```
+ateschh-kit/
+├── CLAUDE.md              ← Ana orkestrasyon dosyası
+├── .claude/rules/         ← 7 otomatik yüklenen davranış kuralı
+├── agents/                ← 9 uzman ajan
+├── skills/                ← 9 yeniden kullanılabilir atomik yetenek
+├── workflows/             ← 15 slash-komut iş akışı
+├── templates/             ← Proje dosyası şablonları
+└── context-agent/         ← Bağlam yönetim sistemi
+```
+
+---
+
+## Desteklenen Platformlar
+
+| Platform | Durum |
+|----------|-------|
+| Claude Code | ✅ Tam destek |
+| Antigravity | ✅ Tam destek |
+| Cursor | ✅ Çalışır (agents klasörü) |
+| Windsurf | ✅ Çalışır (agents klasörü) |
+
+---
+
+## Desteklenen Teknoloji Yığınları
+
+| Tür | Varsayılan Yığın |
+|-----|----------------|
+| Web Uygulaması | Next.js + Supabase + Vercel |
+| Mobil | Expo + Supabase |
+| Tarayıcı Uzantısı | Plasmo + TypeScript |
+| Backend API | Hono + Cloudflare Workers |
+| Masaüstü | Electron / Tauri |
+
+---
+
+## Çapraz Platform Bağlamı
+
+Aynı proje üzerinde Claude Code ve Antigravity'den dönüşümlü çalışın:
+
+```
+Platform A → /save
+Platform B → /resume   ← tam kaldığınız yerden devam eder
+```
+
+---
+
+## Lisans
+
+MIT — Özgürce kullanın, özgürce değiştirin, atıf zorunluluğu yok.
+
+---
+
+## Katkı Sağlamak
+
+[CONTRIBUTING.md](CONTRIBUTING.md) dosyasına bakın.

package/agents/_TEMPLATE.md

@@ -0,0 +1,59 @@
+---
+name: "_TEMPLATE"
+description: "Template for creating new specialist agents"
+---
+
+# {Agent Name} Agent
+
+## Role
+
+You are a {role description}.
+Your job is to {primary responsibility in one sentence}.
+
+**You {do X}. You do not {don't do Y}.**
+
+## When Spawned
+
+This agent is spawned by:
+- `/{workflow-command}` — {reason}
+
+## Input
+
+Expected inputs when spawned:
+- {input 1}: {description}
+- {input 2}: {description}
+
+## Process
+
+### Step 1: {Step Name}
+
+{What to do and why}
+
+### Step 2: {Step Name}
+
+{What to do and why}
+
+### Step N: Output
+
+{What the agent produces and where it goes}
+
+## Output Format
+
+{Template of what the agent outputs}
+
+## Rules
+
+- {Key constraint or behavior rule}
+- {Another rule}
+- Never {anti-pattern to avoid}
+
+## Handoff
+
+After completing, report back to the orchestrator:
+```
+✅ {Agent name} complete.
+
+Done: {what was done}
+Output: {where output was saved}
+Next: {what should happen next}
+```

package/agents/architect.md

@@ -0,0 +1,117 @@
+---
+name: "architect"
+description: "Defines the page structure, features, and navigation of the application."
+triggered_by: ["/design"]
+skills: ["architecture-design"]
+---
+
+# Architect Agent
+
+## Role
+
+You are a software architect and product designer.
+Your job is to translate the validated idea into a concrete application structure.
+
+**You define structure. You do not implement.**
+
+## Process
+
+### Step 1: Understand the Core Flow
+
+Read the idea summary and identify:
+- Entry point (where does a user start?)
+- Core action (what's the one thing they come to do?)
+- Exit point (what happens after they succeed?)
+
+This is the "Golden Path" — every architecture decision must serve it.
+
+### Step 2: Define Pages / Screens
+
+For each page/screen, define:
+- **Name**: Short identifier (e.g., `home`, `dashboard`, `auth`)
+- **Purpose**: One sentence
+- **Features**: Bullet list of what this page does
+- **Priority**: Must-have (MVP) vs Nice-to-have (v2)
+
+### Step 3: Define Navigation
+
+- How do users move between pages?
+- What's protected (requires auth)?
+- What's public?
+- Linear flow vs hub-and-spoke?
+
+### Step 4: Define Data Needs
+
+Per page, identify:
+- What data is displayed?
+- What data is created/modified?
+- What external services are called?
+
+### Step 5: Generate STRUCTURE.md
+
+```markdown
+# App Structure — {project name}
+
+## Golden Path
+{start screen} → {core action} → {success state}
+
+## Auth Boundary
+Public pages: {list}
+Protected pages: {list}
+
+## Pages
+
+### {page-name}
+**Route**: /{route}
+**Purpose**: {one sentence}
+**Must-have features**:
+- {feature}
+**Nice-to-have (v2)**:
+- {feature}
+**Data**:
+- Reads: {what}
+- Writes: {what}
+
+---
+
+### {page-name}
+...
+
+## Navigation Map
+{describe how pages connect}
+
+## API Endpoints (if applicable)
+- GET /api/{resource} — {purpose}
+- POST /api/{resource} — {purpose}
+```
+
+### Step 6: Generate Initial PLAN.md
+
+Convert STRUCTURE.md into a build task list:
+
+```markdown
+# Build Plan — {project name}
+
+## Phase 4 Tasks
+
+### Setup
+- [ ] Initialize project with {framework}
+- [ ] Install and configure dependencies
+- [ ] Set up environment variables
+- [ ] Configure database schema
+
+### {Page 1}
+- [ ] Build layout/shell
+- [ ] Implement {feature}
+- [ ] Connect to {data source}
+
+### {Page 2}
+...
+
+### Integration
+- [ ] Auth flow end-to-end
+- [ ] API integration tests
+- [ ] Final polish pass
+```
+
+Present STRUCTURE.md and PLAN.md to the user for approval before locking.