{
  "schemaVersion": "2.0",
  "updated": "2026-10-08",
  "company": {
    "name": "MEW Monster",
    "nameKo": "뮤몬스터",
    "domain": "mew.monster",
    "country": "South Korea",
    "summary": "Software company building operations software for music academies, a specification and QA platform for product teams, public-data tooling, and AI education services. Uses the Claude API to connect its products.",
    "focus": ["Music academy management SaaS", "Product team tooling", "Public open data", "AI education"]
  },
  "products": [
    {
      "id": "monster-saas",
      "name": "Monster SaaS",
      "category": "Academy management SaaS (flagship)",
      "users": ["academy directors", "branch managers", "front-desk staff", "teachers"],
      "value": "One screen for the front desk with today's lessons, attendance, payments and alerts, plus practice rooms, revenue and teacher payroll for every branch.",
      "features": [
        "Operations dashboard with KPIs, today's lessons and alerts",
        "Branch switcher shared across all pages",
        "Student, guardian and teacher directories",
        "Lesson calendar with booking policies per branch",
        "One-tap attendance and make-up lesson tracking",
        "Practice room settings, hourly occupancy grid and conflict checks",
        "Revenue with 12-month trend per branch, manual entries and CSV",
        "Teacher settlement: calculate, adjust, approve and lock with audit log",
        "Consultation and inquiry inbox"
      ],
      "stack": ["Next.js App Router API on Vite (vinext)", "React 19", "Tailwind CSS 4", "Drizzle ORM", "MySQL 8.4"],
      "screenshots": ["images/saas-settlement.jpg", "images/saas-revenue.jpg", "images/saas-rooms.jpg", "images/saas-inquiries.jpg", "images/saas-students.jpg", "images/saas-calendar.jpg"]
    },
    {
      "id": "devops",
      "name": "DevOps",
      "category": "Feature specification and project operations platform",
      "users": ["product planners", "developers", "designers", "QA", "project leads"],
      "value": "One workspace per project for screens, user flows, event specs, data model, QA test design, issues, documents, schedules, change history and API health.",
      "features": [
        "Linked screens, user flows, event specs and data model",
        "Auto-laid-out flow map with Markdown export",
        "QA test design with draft, approve and retire lifecycle",
        "Spreadsheet-style issue tracking",
        "Documents and file library",
        "Monthly calendar, task sheet and WBS import",
        "API health monitoring with incidents and reports",
        "History with revisions, baselines and compare",
        "Company SSO (OIDC + PKCE) and project roles"
      ],
      "metrics": { "httpEndpoints": 156, "mysqlTables": 58, "unitAndIntegrationTests": 1665, "browserE2eTests": 287 },
      "stack": ["React 19", "Vite", "Mantine", "NestJS 11", "Prisma", "MySQL 8.4", "MinIO/S3", "Docker"],
      "screenshots": ["images/ops-er.jpg", "images/ops-screens.jpg", "images/ops-map.jpg", "images/ops-tests.jpg", "images/ops-global.jpg", "images/ops-schedule.jpg"]
    },
    {
      "id": "api-checker",
      "name": "API Checker (API 확인소)",
      "category": "Public open data tool",
      "users": ["educators", "program teams", "citizen developers"],
      "value": "Shows which of the 10,086 open APIs on Korea's Digital Convergence Platform actually return data, and which work inside the Modu AI Lab vibe-coding studio, with reasons.",
      "features": [
        "Complete catalog of 10,086 APIs in 18 categories",
        "Verification that separates real data from empty or error responses",
        "Studio data tags: restaurants, parking, tourism, weather, bus",
        "Local-only operation; keys kept in memory"
      ],
      "metrics": { "apisCataloged": 10086, "categories": 18, "verificationRecords": 137, "confirmedWithRealData": 106 },
      "screenshots": ["images/dcp-main.jpg"]
    },
    {
      "id": "modu-ai-lounge",
      "name": "Modu AI Lounge (모두의 AI 라운지)",
      "category": "Public website for AI education",
      "users": ["citizens", "families", "students", "educators"],
      "value": "Find hands-on AI classes at five Gyeonggi-province media centers: Gimpo, Goyang, Seongnam, Hwaseong and Anseong.",
      "features": ["Interactive home", "Course finder with 50 courses", "Media center pages", "News with 42 notices", "SEO prerendering and accessibility checks"],
      "screenshots": ["images/lounge-home.jpg", "images/lounge-learn.jpg", "images/lounge-centers.jpg", "images/lounge-news.jpg"]
    },
    {
      "id": "ai-lab-lectures",
      "name": "AI Lab Lectures (모두의 AI 실험실 강의)",
      "category": "Education service",
      "users": ["non-developers", "teachers", "media center learners"],
      "value": "Beginner lectures for the national Modu AI Lab platform, built around one running example: a to-do app, a weather API and an automated morning reminder.",
      "features": ["Vibe coding lectures 1-3", "Public data API lecture", "n8n automation", "Cloud VM and managed dev tools", "Privacy-mask Chrome extension for recording"],
      "screenshots": ["images/lecture-1.jpg", "images/lecture-2.jpg", "images/lecture-3.jpg"]
    }
  ],
  "claudeIntegration": {
    "name": "Monster AI, powered by Claude",
    "summary": "Claude API connects Monster SaaS and DevOps. Claude reads Monster SaaS feature docs, routes, schema and aggregated data, then writes drafts into DevOps for a person to approve, and helps academy staff with summaries and reply drafts.",
    "model": "claude-opus-5-5",
    "apiFeatures": ["structured outputs", "tool use", "prompt caching", "Message Batches", "effort control"],
    "useCases": [
      "Spec drafts from feature docs",
      "Spec drift check against real routes and tables",
      "QA test case drafts",
      "Incident summaries from health checks",
      "Release notes from history compare",
      "Academy insights from aggregated data",
      "Reply drafts for parent inquiries",
      "Summaries of API verification results"
    ],
    "guardrails": ["drafts only, people approve", "typed output validation", "access through DevOps API with roles", "no personal data in prompts", "caching and batches for cost", "per-call usage logging"]
  },
  "privacyNote": "Screenshots use demo data; personal and project details are blurred."
}
