{ "markdown": "
\n\n\"AuraKit\n\n
\n\n[![Version](https://img.shields.io/npm/v/%40smorky85/aurakit?style=flat-square&color=a855f7&label=AuraKit)](https://www.npmjs.com/package/@smorky85/aurakit)\n[![Claude Code](https://img.shields.io/badge/Claude_Code-Native_Skill-7c3aed?style=flat-square&logo=anthropic&logoColor=white)](https://claude.ai/code)\n[![License: MIT](https://img.shields.io/badge/license-MIT-22c55e?style=flat-square)](LICENSE)\n[![Stars](https://img.shields.io/github/stars/smorky850612/Aurakit?style=flat-square&color=f59e0b&logo=github)](https://github.com/smorky850612/Aurakit/stargazers)\n[![npm monthly](https://img.shields.io/npm/dm/%40smorky85/aurakit?style=flat-square&color=cb3837&label=installs%2Fmo&logo=npm)](https://www.npmjs.com/package/@smorky85/aurakit)\n[![npm total](https://img.shields.io/npm/dt/%40smorky85/aurakit?style=flat-square&color=cb3837&label=total%20installs&logo=npm)](https://www.npmjs.com/package/@smorky85/aurakit)\n[![Node](https://img.shields.io/badge/node-%3E%3D18-339933?style=flat-square&logo=node.js&logoColor=white)](package.json)\n\n
\n\n

46 Modes  ·  10 Hooks  ·  23 Agents  ·  6-Layer Security  ·  8 Languages  ·  ~55% Token Savings

\n\n

\nBefore & After  ·  \nQuick Start  ·  \n46 Modes  ·  \nTiers  ·  \nPipeline  ·  \nSecurity  ·  \nNew in v6.5  ·  \nWhy AuraKit  ·  \nFAQ\n

\n\n
\n\n---\n\n## What is AuraKit?\n\n**One command. Full-stack app. Production-ready.**\n\nAuraKit is a [Claude Code](https://claude.ai/code) skill that replaces 20+ manual instructions with a single `/aura` command. It auto-detects what you need, scans your project, generates code with security checks on every file, and commits — all in one shot.\n\n```bash\nnpx @smorky85/aurakit # Install once (~30 seconds, auto-installs jq)\n/aura build: login with JWT # That's it. AuraKit handles the rest.\n```\n\n> [!TIP]\n> **30-second install** → `npx @smorky85/aurakit` or `bash install.sh` (v2.0), then type `/aura` in any project. jq, Python, and git are auto-detected and installed if missing.\n\n---\n\n## 🔄 Before & After\n\n\n\n\n\n\n\n\n\n\n
❌ Without AuraKit✅ With AuraKit
\n\n```\nYou: \"Build a login API\"\nClaude: *generates code*\nYou: \"Wait, add input validation\"\nClaude: *regenerates*\nYou: \"You forgot error handling\"\nClaude: *patches*\nYou: \"Check for SQL injection\"\nClaude: *patches again*\nYou: \"Now write tests\"\nClaude: *generates tests*\nYou: \"The types are wrong...\"\n(30 min later, still going)\n```\n\n\n\n```bash\n/aura build: login with JWT\n\n# AuraKit automatically:\n# → Scans your project stack (Scout/Haiku)\n# → Plans file-by-file build order\n# → Generates with SEC-01~15 rules\n# → Validates types + security + tests\n# → Commits: feat(auth): add login\n# Done. One command. ~3 minutes.\n```\n\n
\n\n---\n\n## ⚡ Quick Start\n\n**1 — Install** (choose one)\n\n```bash\n# macOS (Homebrew)\nbrew install smorky85/tap/aurakit\n\n# All platforms (npm)\nnpx @smorky85/aurakit\n\n# One-liner (macOS / Linux)\ncurl -fsSL https://raw.githubusercontent.com/smorky850612/Aurakit/main/install.sh | bash\n\n# Windows\nnpx @smorky85/aurakit\n\n# From source\nbash install.sh\n```\n\n> [!NOTE]\n> All install methods auto-activate L3~L5 security hooks. **install.sh v2.0** auto-detects and installs: `jq` (winget/scoop/choco/brew/apt/dnf/yum/apk/pacman), checks Node.js, Python, and git. Configures settings.json via Python-first (jq fallback). Installs the AuraKit Nexus status bar.\n\n**2 — Use**\n\n```bash\n# Recommended for daily use (hooks enforce security without per-action dialogs)\nclaude --dangerously-skip-permissions\n\n/aura build: login with JWT # BUILD mode (English)\n/aura 로그인 기능 만들어줘 # BUILD mode (Korean · auto-detect)\n/aura fix: TypeError in auth.ts # FIX mode\n/aura 코드 정리해줘 # CLEAN mode\n/aura Vercel 배포 설정해줘 # DEPLOY mode\n/aura 코드 리뷰해줘 # REVIEW mode\n/aura! 버튼 색상 변경 # QUICK mode · ~60% fewer tokens\n```\n\n> [!WARNING]\n> **What `--dangerously-skip-permissions` means:** Claude won't ask for confirmation on each tool use. This is intentional — AuraKit's hooks (`bash-guard.js`, `security-scan.sh`) replace per-action dialogs with automated enforcement. **Without `install.sh`, security relies only on L1/L2 (agent role isolation + tool blocklist).**\n>\n> **Safer first-time alternative:** Run `claude` without the flag. Claude will ask for permission on each Write/Edit/Bash call.\n\n> [!IMPORTANT]\n> Security rules in `~/.claude/rules/aurakit-security.md` are **always active** — applied to every Claude Code session automatically, even without running `/aura`.\n\n**3 — Update** (existing users)\n\n```bash\n# Recommended — always fetches the latest version from npm\nnpx @smorky85/aurakit@latest\n\n# From source (if you cloned the repo)\ngit pull origin main && bash install.sh\n\n# One-liner (macOS / Linux)\ncurl -fsSL https://raw.githubusercontent.com/smorky850612/Aurakit/main/install.sh | bash\n```\n\n> [!TIP]\n> Running the update command re-runs `install.sh` automatically — hooks, skills, and security rules are all refreshed in place. Your `.aura/` project data (memory, instincts, snapshots) is **never overwritten**.\n\n---\n\n## 🧬 DNA — 8 Core Principles\n\n> AuraKit enforces these 8 principles in every mode, every turn, every output. Any response that violates them is not AuraKit.\n\n
\n\n| # | Principle | Guarantee | Mechanism |\n|:-:|:----------|:----------|:----------|\n| **1** | ⚡ **FAST** | Faster than any skill | Session cache · ConfigHash · QUICK mode · Progressive Load |\n| **2** | ✨ **FLASHY** | Most informative CLI output | StatusLine · Next Actions · Token Report · Pipeline display |\n| **3** | 🔐 **SECURE** | Genuinely top-tier security | 6-layer gates · 10 hooks · SEC-01~15 · bash-guard · security-scan |\n| **4** | 💰 **THRIFTY** | Max token savings even on Opus | Tiered Model · Fail-Only output · Progressive Load · Session cache |\n| **5** | ♾️ **IMMORTAL** | Survives context loss | 65% compact guard · Snapshots · PostCompact restore · Session resume |\n| **6** | 🧠 **EVOLVING** | Gets smarter with every use | Instinct learning · instinct:evolve · Pattern sharing |\n| **7** | 🌐 **UNIVERSAL** | Any platform, any language | 8 languages · 46 modes · 23 agents · 5 platforms · Non-dev QUICK mode |\n| **8** | 🏆 **TOP-TIER** | Best skill, no comparison | Sum of the above 7 |\n\n
\n\n---\n\n## 🏆 Why AuraKit?\n\n> \"I can just prompt Claude myself\" — Yes, but you'll repeat the same 20 instructions every session.\n\n| | Manual Prompting | CLAUDE.md File | **AuraKit v6.5.0** |\n|:---|:---:|:---:|:---:|\n| Security enforcement | Hope for the best | Rules, no enforcement | **10 hooks enforce at write-time** |\n| Context survival | Lost on compact | Partial | **Snapshot + PostCompact auto-restore** |\n| Token efficiency | Wasteful | Manual | **~55% ECO · ~75% MAX (estimated)** |\n| Code review | Manual | Manual | **4 agents in parallel** |\n| Multi-language | English only | English only | **8 languages · 56+ commands** |\n| Learns over time | Starts fresh | Starts fresh | **Instinct engine auto-saves patterns** |\n| jq / tools | Manual setup | Manual setup | **Auto-installed on first run** |\n| Install time | — | 30 min writing rules | **~30 seconds** |\n\n---\n\n## 🎯 46 Modes\n\nAuraKit detects your intent from natural language. Use a namespace prefix (`build:`, `fix:`) when the mode is ambiguous.\n\n### 5 Core Modes — covers 90% of daily use\n\n| Mode | Invoke | What It Does |\n|:-----|:-------|:-------------|\n| **BUILD** | `/aura build: ...` or just describe it | Discovery → micro-plan → generate → triple verify → commit |\n| **FIX** | `/aura fix: ...` or paste the error | Root-cause analysis → minimal change → verify |\n| **REVIEW** | `/aura review:` | 4 parallel agents → VULN-NNN report, A–F grade |\n| **CLEAN** | `/aura clean:` | Dead code removal, 250-line splits, deduplication |\n| **DEPLOY** | `/aura deploy:` | Framework detect → env setup → deploy config → security recheck |\n\n
\n📋 40 Extended Modes — Quality · Planning · Platform · Utility · Autopus-ADK · Auto\n\n
\n\n**Quality & Testing**\n\n| Mode | Trigger | What It Does |\n|:-----|:--------|:-------------|\n| **GAP** | `gap:`, match rate | Design ↔ implementation gap analysis (Match Rate %) |\n| **ITERATE** | `iterate:`, auto-fix | Auto-improve until Match Rate ≥ 90% (max 5 cycles) |\n| **TDD** | `tdd:`, test-first | 🔴 RED → 🟢 GREEN → 🔵 REFACTOR · coverage ≥ 70–90% |\n| **QA** | `qa:`, docker logs | Zero-Script QA via real Docker log analysis |\n| **QA:E2E** | `qa:e2e:setup` | Playwright E2E — auth / CRUD / responsive / CI pipeline |\n| **DEBUG** | `debug:`, 5-why | 4-phase systematic debugging with root-cause tracing |\n\n**Planning & Design**\n\n| Mode | Trigger | What It Does |\n|:-----|:--------|:-------------|\n| **PM** | `pm:`, PRD, discovery | OST + JTBD + Lean Canvas + PRD · 5 PM agents in parallel |\n| **PLAN** | `plan:`, 계획 | Structured plan → `.aura/docs/plan-*.md` |\n| **DESIGN** | `design:`, DB 설계 | DB + API + UI workers parallel → cross-consistency check |\n| **REPORT** | `report:`, 완료 보고서 | 4-perspective value report (user/biz/tech/ops) |\n| **PIPELINE** | `pipeline:`, 개발 순서 | 9-phase guide: Starter / Dynamic / Enterprise |\n| **BRAINSTORM** | `brainstorm:`, HMW | HMW + priority matrix → actionable ideas |\n\n**Advanced Operations**\n\n| Mode | Trigger | What It Does |\n|:-----|:--------|:-------------|\n| **ORCHESTRATE** | `orchestrate:`, leader | Leader / Swarm / Council / Watchdog multi-agent patterns |\n| **BATCH** | `batch:[A,B,C]` | Up to 5 features in parallel Git Worktrees |\n| **LOOP** | `batch:loop:` `until:pass` | Autonomous iteration loop until condition met |\n| **FINISH** | `finish:`, squash | Branch squash merge + Worktree cleanup |\n| **ARCHIVE** | `archive:`, archive:list | Archive features without deleting |\n\n**Platform Specialists**\n\n| Mode | Trigger | What It Does |\n|:-----|:--------|:-------------|\n| **MOBILE** | `mobile:`, react native | React Native / Expo specialized pipeline |\n| **DESKTOP** | `desktop:`, electron | Electron / Tauri specialized pipeline |\n| **BAAS** | `baas:`, supabase | Supabase / Firebase / bkend integration guide |\n| **PAYMENT** | `payment:`, stripe, polar, toss | Stripe · LemonSqueezy · Polar · TossPayments · StepPay — full subscription billing pipeline **(default PRO tier)** |\n\n**Intelligence & Configuration**\n\n| Mode | Trigger | What It Does |\n|:-----|:--------|:-------------|\n| **INSTINCT** | `instinct:show` | View / manage / evolve learned project patterns |\n| **LANG** | `lang:python`, `lang:go` | Force language-specific code reviewer (10 languages) |\n| **MCP** | `mcp:setup`, `mcp:list` | Install & configure 14 MCP server types |\n| **CONTENT** | `content:`, 블로그 | Blog, market research, IR deck, tech docs, email, social |\n| **STATUS** | `status`, 현재 상태 | Current work state from `.aura/snapshots/` |\n| **STATUS:HEALTH** | `status:health` | Health Dashboard — Match Rate · security score · coverage · Tech Debt |\n| **CONFIG** | `config:set` | Manage `.aura/config.json` settings |\n\n**Utility**\n\n| Mode | Trigger | What It Does |\n|:-----|:--------|:-------------|\n| **STYLE** | `style:`, learning | Switch output persona: learning / expert / concise |\n| **SNIPPETS** | `snippets:`, 스니펫 | Save and reuse prompt templates |\n| **QUICK (`!`)** | `/aura! request` | Protocol-minimal, single file, ~60% token savings |\n| **BUILD_RESOLVER** | *(auto on V1 fail)* | Language-specific build error resolver (7 languages) |\n\n**Autopus-ADK Modes** *(absorbed v6.5)*\n\n| Mode | Trigger | What It Does |\n|:-----|:--------|:-------------|\n| **SPEC** | `spec:new`, `spec:` | EARS-format spec → `spec.md` + `acceptance.md` (Given/When/Then) |\n| **LORE** | `lore:commit`, `lore:` | 9-trailer decision commit (Constraint/Rejected/Confidence/…/Related) |\n| **ANNOTATE** | `annotate:`, `ax:` | @AX inline annotations (NOTE/WARN/ANCHOR/TODO · `[AUTO]` prefix · CYCLE tracking) |\n| **EXPERIMENT** | `experiment:`, `xloop:` | XLOOP autonomous experiment loop (measure→change→decide, circuit breaker N=10) |\n| **EXPLAIN** | `explain:`, 어떻게 동작 | Step-by-step code explanation (learning-first) |\n| **ROLLBACK** | `rollback:`, undo | Safe revert with git status confirmation before reset |\n| **MIGRATE** | `migrate:`, upgrade | Version migration pipeline — dependency upgrades + breaking change handling |\n| **ESCALATE** | `escalate:` | Promote current task to Opus sub-agent · auto-return to base tier after |\n\n
\n\n---\n\n## 📊 Quality Tiers\n\n| Tier | Invoke | Scout | Builder | Reviewer | TestRunner | Savings (est.) |\n|:-----|:-------|:------|:--------|:---------|:-----------|:---------------|\n| **QUICK** | `/aura! request` | — | Sonnet | — | — | ~60% |\n| **ECO** *(default)* | `/aura request` | Haiku | Sonnet | Sonnet | Haiku | ~55% |\n| **PRO** | `/aura pro request` | Haiku | **Sonnet+Amplifier v2** | Sonnet | Haiku | ~20% |\n| **MAX** | `/aura max request` | Sonnet | **Opus** | **Opus** | Sonnet | ~0% |\n\n- **QUICK** — Color changes, text edits, single-file tweaks\n- **ECO** — Feature development, most daily work *(recommended)*\n- **PRO** — Auth, payments, complex business logic (Sonnet+Amplifier v2, Opus removed)\n- **MAX** — Security audits, architecture design, production-critical features\n\n> [!NOTE]\n> All token savings figures are **estimates** based on tier routing and context load reduction. Context load reduction (v5.1 82KB → v6 20KB) is **measured**. Independent benchmarks (like Aider's Polyglot 64%) do not exist for AuraKit.\n\n---\n\n## ⚙️ How It Works\n\n```mermaid\nflowchart TD\n A([\"/aura request\"]) --> B[\"Mode Detection\\n46 modes · auto from natural language\"]\n B --> C[\"Scout Agent · Haiku\\nProject scan → .aura/project-profile.md\\nConfigHash: skip if unchanged\"]\n C --> D[\"Micro-Plan\\n200-token file-by-file build order\\nInstinct patterns loaded\"]\n D --> E[\"Builder Agent\\nECO: Sonnet · PRO: Sonnet+Amp v2 · MAX: Opus\\nSEC-01~15 + Language Reviewer\"]\n E --> F[\"Security Hooks\\nbash-guard.js · security-scan.sh\\nL3 + L5 enforcement\"]\n F --> G{\"Triple Verify\\n(parallel)\"}\n G --> H[\"V1 · Build / Type Check\"]\n G --> I[\"V2 · Security + Review · Sonnet\"]\n G --> J[\"V3 · TestRunner · Haiku\"]\n H --> K[\"Snapshot Checkpoint\\n.aura/snapshots/current.md\\nInstinct auto-save\"]\n I --> K\n J --> K\n K --> |\"More files\"| E\n K --> |\"Complete\"| L([\"git commit\\nNext Actions report\\nToken savings report\"])\n```\n\n---\n\n## 🔐 6-Layer Security\n\nEvery generated file passes through up to 6 security gates.\n\n
\n\n| Layer | Name | What It Checks | Active Without install.sh? |\n|:-----:|:-----|:---------------|:--------------------------:|\n| **L1** | Agent Roles | Per-agent read/write boundaries in system prompts | ✅ Always |\n| **L2** | Disallowed Tools | Write/Edit/Bash blocklist for read-only agents | ✅ Always |\n| **L3** | Bash Guard | `rm -rf`, `DROP TABLE`, `eval`, dangerous shell commands | ❌ Requires install.sh |\n| **L4** | Worktree Isolation | Agent subprocesses run in isolated Git Worktrees | ❌ Requires install.sh |\n| **L5** | Security Scan | API keys, hardcoded secrets, SQL injection, XSS, JWT in localStorage | ❌ Requires install.sh |\n| **L6** | Dependency Audit | `npm audit --audit-level=high` on BUILD and FIX | ✅ Auto in pipeline |\n\n
\n\n> [!WARNING]\n> **L3, L4, L5 are only active after running `install.sh` or `npx @smorky85/aurakit`.** Without installation, only L1 (agent role boundaries) and L2 (tool blocklist) protect you. If you skip installation, omit `--dangerously-skip-permissions` so manual confirmation acts as your safety net.\n\n**Automatically blocked by SEC-01~15:**\n\n```\nlocalStorage.setItem('token', ...) → httpOnly Cookie required (SEC-02)\nRaw SQL string concatenation → parameterized queries (SEC-01)\neval() / exec() with user input → blocked (SEC-05)\nHardcoded API keys / passwords → blocked (SEC-03)\n.env not in .gitignore → commit blocked (SEC-10)\nMath.random() for security tokens → crypto.randomBytes() (SEC-08)\nHTTP fallback in external calls → HTTPS only (SEC-09)\n```\n\n
\n🔒 Full SEC-01~15 OWASP Reference\n\n
\n\n```\nSEC-01 SQL injection prevention SEC-09 HTTPS-only external calls\nSEC-02 Auth tokens (httpOnly Cookie) SEC-10 .gitignore enforcement\nSEC-03 Secret management SEC-11 NoSQL/Command/XML/LDAP injection\nSEC-04 Input validation (Zod/Pydantic/@Valid) SEC-12 Cryptography (AES-256+)\nSEC-05 eval/exec blocking SEC-13 Dependency audit\nSEC-06 Error info leakage prevention SEC-14 Security event logging\nSEC-07 CORS whitelist enforcement SEC-15 SSRF prevention\nSEC-08 Secure random (crypto module)\n```\n\nAll 15 rules are enforced both inline (code generation) and at runtime (security-scan.sh hook on every Write/Edit).\n\n
\n\n---\n\n## ✨ New in v6.5\n\n
\n🤖 Autopus-ADK Absorption — 16 Specialized Agents + 4 New Modes\n\n
\n\nv6.5 fully absorbs the Autopus-ADK framework (\"A harness of the agents, by the agents, for the agents\"):\n\n**16 New Specialized Agents:**\n`planner` · `executor` · `tester` · `validator` · `reviewer` · `security-auditor` · `annotator` · `architect` · `debugger` · `deep-worker` · `devops` · `explorer` · `frontend-specialist` · `perf-engineer` · `spec-writer` · `ux-validator`\n\n**4 New Modes:**\n\n| Mode | Command | What It Does |\n|:-----|:--------|:-------------|\n| **SPEC** | `/aura spec:new` | EARS-format spec with Given/When/Then acceptance criteria |\n| **LORE** | `/aura lore:commit` | 9-trailer decision tracking commit for architectural decisions |\n| **ANNOTATE** | `/aura annotate:` | @AX inline code annotations (NOTE/WARN/ANCHOR/TODO) |\n| **EXPERIMENT** | `/aura experiment:` | XLOOP autonomous experiment loop with circuit breaker (N=10) |\n\n**Key Concepts Absorbed:**\n- RALF loop: RED→GREEN→REFACTOR→LOOP (max 3 retries)\n- TRUST 5 review: Tested/Readable/Unified/Secured/Trackable\n- Lead/Builder/Guardian agent team topology\n- ICE scoring + SCAMPER + MoSCoW (brainstorming)\n- Worktree isolation R1-R7 (3-check conflict detection)\n- Behavioral assertion rule (NoError 단독 금지 — observable behavior required)\n\n
\n\n
\n✅ Claude Code Compliance — Hook cleanup · SKILL.md -105 lines · Resource splits\n\n
\n\nv6.5 brings AuraKit into full compliance with Claude Code official guidelines:\n\n| Change | Before | After |\n|:-------|:-------|:------|\n| Hook events | 16 events (incl. `TeammateIdle` which CC doesn't emit) | **10 valid CC events only** |\n| SKILL.md size | 662 lines | **557 lines** (-105 lines) |\n| `build-resolvers.md` | 1,352 lines (monolithic) | **41-line router + 7 language files** |\n| `payment-pipeline.md` | 1,043 lines (monolithic) | **251-line router + 5 provider files** |\n| PRO tier model | Opus | **Sonnet+Amplifier v2** (Opus removed from PRO) |\n\n- **TeammateIdle removed** — not a real Claude Code event; agent memory now saved via `SubagentStop` hook\n- **Progressive Load** — resource files split by language/provider so only the needed file loads per task\n\n
\n\n---\n\n## ✨ New in v6.4\n\n
\n💳 PAYMENT Mode — Stripe · LemonSqueezy · Polar · TossPayments · StepPay\n\n
\n\nv6.4 adds a dedicated payment pipeline for subscription billing — the most security-critical feature in any SaaS:\n\n```bash\n/aura payment: Stripe 구독 결제 붙여줘 # 한국어\n/aura payment: add Stripe subscription billing # English\n/aura payment: TossPayments 정기결제 연동 # 한국 서비스\n```\n\n| Provider | Market | Tax Handling | Subscription | Recommended for |\n|:---------|:-------|:-------------|:-------------|:----------------|\n| **Stripe** | Global | Manual (Stripe Tax) | ✅ Full | Global SaaS |\n| **LemonSqueezy** | Global | ✅ Auto (MoR) | ✅ Full | Digital products |\n| **Polar** | Global | ✅ Auto (MoR) | ✅ | Open source |\n| **TossPayments** | 🇰🇷 Korea | ❌ | ✅ Regular billing | 국내 서비스 |\n| **StepPay** | 🇰🇷 Korea | ❌ | ✅ | 한국 BNPL |\n\n**What it builds automatically:**\n- DB schema: `subscription_plans` + `subscriptions` + `webhook_events`\n- Checkout → subscription create → billing portal flow\n- Webhook handler with **signature verification** + idempotency (`INSERT ON CONFLICT`)\n- Subscription middleware for route access control\n- `.env.example` with all required keys\n\n> ⚠️ **Default tier: PRO** (Sonnet+Amplifier v2) — payment code is business-critical. Security audit (VULN-001~012) runs automatically.\n\n
\n\n
\n🚀 install.sh v2.0 — Zero-friction setup\n\n
\n\nv6.3 completely rewrites the installer with intelligent dependency management:\n\n| Feature | v6.2 | v6.3 |\n|:--------|:-----|:-----|\n| jq installation | Warn and exit if missing | **Auto-install** (winget/scoop/choco/brew/apt/dnf/yum/apk/pacman) |\n| Settings update | jq required | **Python-first** (jq fallback — works without jq) |\n| OS detection | None | `uname -s` → Darwin / Linux / Windows (MINGW/MSYS/CYGWIN) |\n| Node.js check | None | **Hard check** — exits with install link if missing |\n| Python check | None | **Soft check** — warns if missing (statusline gracefully degrades) |\n| git check | None | **Soft check** — warns if missing |\n| Language filter | All languages | `--lang=ko,en,jp,zh,es,fr,de,it` selective install |\n| Status bar | Not installed | **AuraKit Nexus** auto-installed |\n\n```bash\nbash install.sh # Full install (all languages)\nbash install.sh --lang=en,ko # English + Korean only\n```\n\n
\n\n
\n📊 AuraKit Nexus Status Bar — Real-time token & session info\n\n
\n\ninstall.sh v2.0 installs the AuraKit Nexus status bar automatically:\n\n```\n💰 ECO | ctx: 23% | ↑12.4K ↓8.1K = 20.5K (7회) | 주간: 143K | /aura review\n```\n\n- **Subscription users** → daily/weekly remaining % (일↓88% 주↓92%)\n- **API users** → actual cost display ($0.23)\n- **Auto-resizes**: 3-line (≥80 cols) · 2-line · 1-line (<55 cols)\n- **Language auto-detect**: 한국어/日本語/中文/English/...\n\n
\n\n
\n🪝 10 Hook Events — Core lifecycle automation\n\n
\n\nAuraKit registers 10 valid Claude Code hook events covering the full agent lifecycle:\n\n| Hook Event | Handler File | Function |\n|:-----------|:-------------|:---------|\n| `SessionStart` | `pre-session.sh` | `.env` security · package manager detect · snapshot check |\n| `UserPromptSubmit` | `korean-command.js` | IME reverse-transliteration routing |\n| `UserPromptSubmit` | `auradkit-detect.js` | Autopus-ADK detect + routing |\n| `PreToolUse` | `security-scan.sh` | Secret pattern detection **(L5)** |\n| `PreToolUse` | `bash-guard.js` | Dangerous command blocking **(L3)** |\n| `PostToolUse` | `build-verify.sh` | Compile / type-check after every file |\n| `PostToolUse` | `bloat-check.sh` | 250-line split warning |\n| `PostToolUse` | `instinct-auto-save.js` | Instinct pattern auto-save |\n| `PostToolUse` | `auto-format.js` | Prettier / gofmt / black / rustfmt |\n| `PostToolUse` | `governance-capture.js` | Architecture decision audit trail |\n| `PostToolUseFailure` | `post-tool-failure.js` | MCP recovery + failure tracking |\n| `Stop` | `session-stop.js` | Session metrics · Instinct hints · incomplete task alert |\n| `PreCompact` | `pre-compact-snapshot.sh` | Save context state before compaction |\n| `PostCompact` | `post-compact-restore.sh` | Restore context after compaction |\n| `SubagentStart` | `subagent-start.js` | Agent lifecycle — registration + spawn limit check |\n| `SubagentStop` | `subagent-stop.js` | Agent completion + memory save (replaces TeammateIdle) |\n\n
\n\n
\n🎯 STATUS:HEALTH — Mode 36 — Project Health Dashboard\n\n
\n\n```bash\n/aura status:health\n```\n\nGenerates a real-time project health report:\n\n```\n╔══════════════════════════════════════════════════╗\n║ AuraKit Health Dashboard ║\n╠══════════════════════════════════════════════════╣\n║ Match Rate: █████████░ 87% (target: ≥90%) ║\n║ Security: ██████████ 98% (SEC-01~15) ║\n║ Test Coverage: ████████░░ 78% (target: ≥80%) ║\n║ Tech Debt: ████░░░░░░ Low (12 items) ║\n║ Doc Lifecycle: ██████░░░░ 61% (needs update) ║\n╠══════════════════════════════════════════════════╣\n║ Recommended: /aura iterate → /aura tdd ║\n╚══════════════════════════════════════════════════╝\n```\n\n
\n\n---\n\n## ✨ More v6 Features\n\n
\n🧠 Sonnet Amplifier — Opus-level quality from Sonnet\n\n
\n\nAuraKit v6 makes Sonnet produce Opus-quality code by forcing structured reasoning before every file:\n\n1. **I/O Contract** — Define types, return values, error cases\n2. **Existing Code Check** — Verify import/naming/style compatibility\n3. **Edge Case Discovery** — Side effects? Concurrency? Input boundaries? External failures?\n4. **SEC/Q Rule Selection** — Pick applicable security & quality rules\n5. **Then implement** — Only after all 4 steps\n\nThis eliminates Sonnet's tendency to rush to code, producing measurably better output.\n\n
\n\n
\n🧠 Instinct Learning Engine — AuraKit gets smarter with every session\n\n
\n\nAuraKit learns your project's patterns after each BUILD/FIX and applies them automatically on the next run.\n\n```bash\n/aura instinct:show # All learned patterns\n/aura instinct:show auth # Filter by category\n/aura instinct:prune # Remove low-score patterns\n/aura instinct:evolve # Auto-improve + integrate anti-patterns\n/aura instinct:export # Backup / share with team\n/aura instinct:import team.json # Import from another project\n/aura instinct:reset # ⚠️ Reset all patterns (irreversible)\n```\n\n| Before install.sh | After install.sh |\n|:------------------|:-----------------|\n| Patterns saved by Claude's judgment after BUILD/FIX | `instinct-auto-save.js` triggers on every Write/Edit — **fully automated** |\n| Semi-automated | **Fully automated** |\n\n
\n\n
\n🌐 Language-Specific Code Reviewers — 10 languages with framework awareness\n\n
\n\nAuraKit auto-detects your project language and applies a specialized reviewer in the V2 step.\n\n| Language | Reviewer Focus |\n|:---------|:--------------|\n| TypeScript | Strict null checks, React/Next.js/NestJS patterns, generic type safety |\n| Python | Type hints, async patterns, FastAPI/Django/Flask idioms |\n| Go | Goroutine safety, error wrapping, interface design |\n| Java | Generics, Spring patterns, checked exception handling |\n| Rust | Ownership, lifetimes, unsafe usage review |\n| Kotlin | Coroutines, null safety, idiomatic style |\n| C++ | Memory safety, RAII, smart pointer usage |\n| Swift | ARC, optionals, Combine / async-await patterns |\n| PHP | Type coercion risks, PDO usage, injection vectors |\n| Perl | Modern::Perl, regex safety |\n\n```bash\n/aura lang:python review: # Force Python reviewer\n/aura lang:go fix: goroutine leak\n```\n\n
\n\n
\n🔌 14 MCP Server Configurations\n\n
\n\n```bash\n/aura mcp:setup # Interactive setup wizard\n/aura mcp:list # Show all 14 available configs\n/aura mcp:check # Verify installed servers\n/aura mcp:add playwright # Add a specific server\n```\n\nSupported: Playwright · GitHub · Slack · Linear · Notion · Supabase · PostgreSQL · MongoDB · Redis · Stripe · Vercel · AWS · Sanity · and more.\n\n
\n\n
\n📍 Next Actions System — Auto-suggest after every task\n\n
\n\nAfter every mode completion, AuraKit shows what to do next and reports real token savings:\n\n```\n━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\n📍 Done: BUILD — JWT Login (8 files)\n💰 Token Report:\n Baseline (manual est.): ~18,200 tokens\n Actual used: ~7,800 tokens\n Saved: 57% (10,400 tokens)\n ├─ Discovery effect: -3,200 (삽질 방지)\n ├─ Tier model effect: -4,800 (haiku Scout + sonnet V2)\n ├─ Cache hit: -1,500 (session cache + ConfigHash)\n └─ Instinct reuse: -900 (3 patterns applied)\n📊 Pipeline: ████████░░░░ 4/7\n PM ✓ → PLAN ✓ → DESIGN ✓ → BUILD ✓ → REVIEW → ITERATE → DEPLOY\n🔜 Next: /aura review → /aura deploy\n━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\n```\n\n
\n\n
\n🌐 Cross-Platform (5 Platforms) — Claude Code, Codex, Cursor, Manus, Windsurf\n\n
\n\n| Platform | Support | Setup | Model Mapping |\n|:---------|:-------:|:------|:-------------|\n| **Claude Code** | ✅ Full | Native — no extra setup | haiku / sonnet / opus |\n| **OpenAI Codex CLI** | ✅ Full | SKILL.md auto-recognized | haiku→gpt-4o-mini, sonnet→gpt-4o, opus→o3 |\n| **Cursor** | ✅ Supported | `.cursorrules`, Agent Mode | Cursor model selector |\n| **Manus** | ✅ Supported | System prompt + multi-agent | Manus routing |\n| **Windsurf** | ✅ Supported | `.windsurfrules`, Cascade | Windsurf selector |\n| **Aider** | ⚠️ Partial | `.aider.conf.yml` | BUILD/FIX only |\n| **Gemini CLI** | 🔬 Experimental | System prompt | Unverified |\n\n```bash\nnpx @smorky85/aurakit # Claude Code (default)\nnpx @smorky85/aurakit --platform=codex # Codex CLI adapter\nnpx @smorky85/aurakit --platform=cursor # Cursor adapter\nnpx @smorky85/aurakit --platform=manus # Manus adapter\n```\n\n
\n\n
\n🤖 Dynamic Agent Spawning — With circuit breaker\n\n
\n\nAgents can spawn child agents when tasks are too large. Hard limits prevent runaway:\n\n| Limit | Value |\n|:------|:------|\n| Max depth | 3 (Agent → Child → Grandchild) |\n| Max total per session | 12 |\n| Max concurrent | 5 |\n| Timeout per agent | 5 minutes |\n| Circuit breaker | 3 consecutive failures → freeze + user alert |\n| Token budget | 30% of remaining context per agent |\n\nAgent results auto-saved to `.aura/agent-memory/[agent].json` via `SubagentStop` hook.\n\n
\n\n
\n🔄 Loop Operator — Autonomous iteration until done\n\n
\n\n```bash\n/aura batch:loop:[task] until:pass max:5 # Repeat until tests pass\n/aura batch:loop:[task] until:90% max:3 # Repeat until gap ≥ 90%\n/aura batch:loop:[task] until:no-error max:10 # Repeat until error-free\n```\n\nEach iteration runs in isolation. Loop stops when the condition is met or max is reached.\n\n
\n\n---\n\n## 💰 Token Optimization\n\n> All savings figures are **estimates**. Context load reduction is **measured** (v5.1: 82KB → v6: 20KB = 75.6% reduction).\n\n| Technique | How | Savings (est.) |\n|:----------|:----|:---------------|\n| **SKILL.md slim** | Detailed content delegated to resources, 75% context load reduction | Load savings |\n| **Tiered Model** | Scout / TestRunner use Haiku by default | ~40% |\n| **Fail-Only Output** | Agents return \"Pass\" or errors only — no verbose logging | ~30% |\n| **Progressive Load** | Resource files loaded per-mode only, not all at once | ~10% |\n| **Session Cache (B-0)** | Skip project re-scan if within 2 hours | ~10% |\n| **ConfigHash** | Rescan only when `package.json` / lockfile changes | ~10% |\n| **Graceful Compact** | Triggers at 65% context + checkpoint saves state | waste eliminated |\n| **QUICK Mode** | `/aura!` — no protocol, single file | ~60% |\n\n---\n\n## 🌍 Multilingual Commands\n\n8 languages, 56+ commands. Type in your language without switching input methods.\n\n
\n\n| Language | Start | Build | Fix | Clean | Deploy | Review |\n|:---------|:------|:------|:----|:------|:-------|:-------|\n| 🇺🇸 English | `/aura` | `/aura build:` | `/aura fix:` | `/aura clean:` | `/aura deploy:` | `/aura review:` |\n| 🇰🇷 Korean | `/아우라` | `/아우라빌드` | `/아우라수정` | `/아우라정리` | `/아우라배포` | `/아우라리뷰` |\n| 🇯🇵 Japanese | `/オーラ` | `/オーラビルド` | `/オーラ修正` | `/オーラ整理` | `/オーラデプロイ` | `/オーラレビュー` |\n| 🇨🇳 Chinese | `/奥拉` | `/奥拉构建` | `/奥拉修复` | `/奥拉清理` | `/奥拉部署` | `/奥拉审查` |\n| 🇪🇸 Spanish | `/aura-es` | `/aura-construir` | `/aura-arreglar` | `/aura-limpiar` | `/aura-desplegar` | `/aura-revisar` |\n| 🇫🇷 French | `/aura-fr` | `/aura-construire` | `/aura-corriger` | `/aura-nettoyer` | `/aura-deployer` | `/aura-reviser` |\n| 🇩🇪 German | `/aura-de` | `/aura-bauen` | `/aura-beheben` | `/aura-aufraeumen` | `/aura-deployen` | `/aura-pruefen` |\n| 🇮🇹 Italian | `/aura-it` | `/aura-costruire` | `/aura-correggere` | `/aura-pulire` | `/aura-distribuire` | `/aura-rivedere` |\n\n
\n\n> **v6.3**: All 56+ multilingual commands are handled by the core `/aura` skill via auto-detection. No separate skill folders needed — freeing ~83% of Claude Code's skill description budget.\n\n**IME support**: Korean and Japanese IME reverse-transliteration is handled automatically by `hooks/korean-command.js`.\n\n---\n\n## 🔧 Compatibility\n\n| Tool | SKILL.md | Hooks | Agents | Support |\n|:-----|:--------:|:-----:|:------:|:-------:|\n| **Claude Code** *(recommended)* | ✅ | ✅ 16 handlers · 10 events | ✅ 23 agent types | **Full** |\n| **OpenAI Codex CLI** | ✅ | ✅ sandbox pre/post | ✅ agents.md | **Full** |\n| **Cursor** | ✅ | ⚠️ VS Code Tasks | ✅ Agent Mode | **Supported** |\n| **Manus** | ✅ | ✅ event system | ✅ native multi-agent | **Supported** |\n| **Windsurf** | ✅ | ⚠️ VS Code Tasks | ✅ Cascade | **Supported** |\n| **Aider** | ✅ | ❌ | ❌ | Partial |\n| **Gemini CLI** | ✅ | ❌ | ❌ | Experimental |\n\n---\n\n
\n📁 Full Repository Structure\n\n
\n\n```\naurakit/ # v6.5.0\n├── skills/\n│ ├── aura/ # Main skill — single /aura entry point\n│ │ ├── SKILL.md # Core instructions (~20KB, 46 modes)\n│ │ └── resources/ # 60+ mode-specific pipeline guides\n│ │ ├── build-pipeline.md\n│ │ ├── fix-pipeline.md\n│ │ ├── security-rules.md # SEC-01~15 full reference\n│ │ ├── instinct-system.md\n│ │ ├── language-reviewers.md # 10 language reviewers\n│ │ ├── loop-pipeline.md\n│ │ ├── mcp-configs.md # 14 MCP server configs\n│ │ ├── cross-harness.md # 5-platform guide\n│ │ ├── status-dashboard.md # Health Dashboard\n│ │ ├── build-resolvers.md # Router → language-specific files\n│ │ ├── build-resolvers-go.md · -rust.md · -python.md · -java.md · …\n│ │ ├── payment-pipeline.md # Router → provider-specific files\n│ │ ├── payment-stripe.md · -lemonsqueezy.md · -polar.md · -toss.md · …\n│ │ ├── autopus-spec.md # EARS SPEC system (v6.5)\n│ │ ├── autopus-lore.md # Decision tracking commits (v6.5)\n│ │ ├── autopus-annotate.md # @AX annotation system (v6.5)\n│ │ ├── autopus-pipeline.md # 5-phase pipeline v2 (v6.5)\n│ │ └── ... # +40 more guides\n│ ├── aura-compact/ # Snapshot + /compact shortcut\n│ └── aura-guard/ # Token budget monitor\n├── agents/ # 23 specialized agents\n│ ├── scout.md # Read-only project scanner (Haiku)\n│ ├── worker.md # Reviewer + TestRunner (Sonnet)\n│ ├── gap-detector.md # Design↔implementation gap (Haiku)\n│ ├── security.md # OWASP Top 10 audit (Sonnet)\n│ ├── pm-discovery.md # OST opportunity mapping (Haiku)\n│ ├── pm-strategy.md # JTBD + Lean Canvas (Haiku)\n│ ├── pm-prd.md # PRD generation (Sonnet)\n│ ├── planner.md # Task planning agent (v6.5)\n│ ├── executor.md # Stack-specific implementation (v6.5)\n│ ├── tester.md # Test generation + TDD (v6.5)\n│ ├── validator.md # 8-check Gate 2 validation (v6.5)\n│ ├── reviewer.md # TRUST 5 code review (v6.5)\n│ ├── security-auditor.md # Deep security audit (v6.5)\n│ ├── annotator.md # @AX annotation writer (v6.5)\n│ ├── architect.md # Architecture design (v6.5)\n│ ├── debugger.md # 5-WHY systematic debug (v6.5)\n│ ├── deep-worker.md # Complex long-running tasks (v6.5)\n│ ├── devops.md # CI/CD + infrastructure (v6.5)\n│ ├── explorer.md # Codebase exploration (v6.5)\n│ ├── frontend-specialist.md # UI/UX + accessibility (v6.5)\n│ ├── perf-engineer.md # Performance optimization (v6.5)\n│ ├── spec-writer.md # EARS spec generation (v6.5)\n│ └── ux-validator.md # UX + usability validation (v6.5)\n├── hooks/ # 16 handler files, 10 hook events\n│ ├── lib/\n│ │ ├── common.js # Shared: addContext, allow, block\n│ │ ├── snapshot.js # Snapshot read/write helpers\n│ │ └── python.js # Cross-platform Python executor\n│ ├── security-scan.sh # Secret pattern detection (L5)\n│ ├── bash-guard.js # Dangerous command blocking (L3)\n│ ├── build-verify.sh # Compile / type-check after each file\n│ ├── bloat-check.sh # 250-line split warning\n│ ├── instinct-auto-save.js # Instinct pattern auto-save\n│ ├── auto-format.js # Prettier / gofmt / black / rustfmt\n│ ├── governance-capture.js # Architecture decision audit trail\n│ ├── post-tool-failure.js # MCP recovery + failure tracking\n│ ├── session-stop.js # Session metrics + Instinct hints\n│ ├── subagent-start.js # Agent lifecycle — registration\n│ ├── subagent-stop.js # Agent completion + memory save\n│ ├── korean-command.js # IME reverse-transliteration\n│ ├── auradkit-detect.js # Autopus-ADK detect + routing\n│ ├── pre-compact-snapshot.sh # Save context before compact\n│ └── post-compact-restore.sh # Restore context after compact\n├── rules/\n│ └── aurakit-security.md # Always-active security rules\n├── templates/\n│ ├── design-system-default.md # CSS variable token defaults\n│ └── project-profile-template.md\n├── statusline/ # AuraKit Nexus status bar\n│ ├── statusline-command.sh\n│ └── statusline-parser.py\n├── tests/ # AuraScore test suite\n│ ├── run-tests.sh\n│ ├── test-build-basic.md\n│ ├── test-scout-detect.md\n│ ├── test-instinct-save.md\n│ ├── test-quick-mode.md\n│ └── test-safety-net.md\n├── install.sh # Installer v2.0 (auto-jq, Python-first)\n├── CHANGELOG.md # Version history\n├── SECURITY.md # Vulnerability reporting policy\n├── CONTRIBUTING.md\n└── LICENSE # MIT\n```\n\n
\n\n---\n\n## 🤝 Contributing\n\nWe welcome contributions! Add language reviewers, framework patterns, hooks, or security rules.\n\n```bash\n# Quick contribution workflow\nfork → branch (feat/reviewer-csharp) → create files → bash tests/run-tests.sh → PR\n```\n\nSee [CONTRIBUTING.md](CONTRIBUTING.md) for templates and quality requirements.\n\n| Contribution | Path | Requirements |\n|:-------------|:-----|:-------------|\n| Language Reviewer | `resources/language-reviewers.md` | 10 rules + 5 checklist items + V1 command |\n| Framework Pattern | `resources/framework-patterns.md` | File structure + 10 rules |\n| Hook | `hooks/[name].js` or `.sh` | Error handling + install.sh registration |\n| Security Rule | `rules/aurakit-security.md` SEC-16+ | OWASP/CWE mapping + no duplicates |\n| Mode | `skills/aura/resources/[mode]-pipeline.md` | Full pipeline spec + SKILL.md entry |\n\n---\n\n## ❓ FAQ\n\n
\nClick to expand all questions\n\n
\n\n**What frameworks does AuraKit support?**\n\nAny framework. Scout reads `package.json`, `tsconfig`, `tailwind.config`, `prisma.schema`, `Dockerfile`, `go.mod`, `pyproject.toml`, and more to adapt automatically. Next.js, React, Vue, Svelte, Express, FastAPI, Django, Spring — all work out of the box.\n\n**Does AuraKit work with existing projects?**\n\nYes. AuraKit scans your existing codebase first (via Scout agent), then generates code that matches your existing conventions, styling, and architecture.\n\n**What happens if I already have hooks configured?**\n\n`install.sh` (v2.0) updates your `settings.json` using Python-first logic that preserves your existing configuration while adding AuraKit hooks.\n\n**Does AuraKit work on Windows?**\n\nYes, no WSL required. Shell hooks (`.sh`) run via Git Bash; Node hooks (`.js`) run cross-platform. Git Bash is recommended for Claude Code itself.\n\n**How does compact defense work?**\n\nAuraKit sets `CLAUDE_AUTOCOMPACT_PCT_OVERRIDE=65`, triggering compaction at 65% token usage instead of the default 95%. A `PreCompact` hook saves your current work state. A `PostCompact` hook restores it. Zero context loss.\n\n**Is my code sent anywhere?**\n\nNo. AuraKit is a local skill. Everything runs inside your Claude Code session. No external APIs, no telemetry, no data collection.\n\n**How is AuraKit different from just prompting Claude?**\n\nManual prompting requires re-explaining your project, conventions, and requirements every session. AuraKit pre-loads all of that automatically via Scout, enforces security through hooks, uses specialized agents for each role, and learns your patterns over time — without extra effort after initial install.\n\n**The token savings numbers seem high. Are they real?**\n\nContext load reduction (v5.1 82KB → v6 20KB) is **measured**. Per-task savings (~55% ECO, ~75% MAX) are **estimates** based on tier routing and prompt engineering — no independent benchmark like Aider's Polyglot 64% exists for AuraKit.\n\n**Do I need all 10 hook events? Can I use only some?**\n\nYes. Install registers all hooks, but each is independent. Remove any entry from `settings.json` to disable it. The security hooks (L3/L5) are most critical; the rest are quality-of-life.\n\n
\n\n---\n\n
\n\n
\n\n### Try it now — it takes ~30 seconds:\n\n```bash\nnpx @smorky85/aurakit\n```\n\nThen open any project and type `/aura`.\n\n
\n\n**AuraKit v6.5.0 — 46 modes · 10 hooks · 23 agents · 6-layer security · 5 platforms · ~55% token savings**\n\n
\n\n[![Star on GitHub](https://img.shields.io/github/stars/smorky850612/Aurakit?style=social)](https://github.com/smorky850612/Aurakit/stargazers)\n\n
\n\nIf AuraKit saves you time, [give it a star ⭐](https://github.com/smorky850612/Aurakit/stargazers) — it helps others find it.\n\n
\n\nGitHub  ·  npm  ·  Issues  ·  Contribute  ·  Changelog  ·  MIT License\n\n
\n\n
\n", "bytes": 43554, "sha": "36b1041a00f075790e33446a5344bb04e1181596c9bca6c216b9eacc998beb2d", "repo_slug": "smorky850612/aurakit", "fonte": "repo", "truncated": false, "api": "https://agentalog.com/api/listings/plg_smorky850612_aurakit_aurakit_0beafe4f/readme" }