make-me-unicorn 0.5.0

Stop building blind. The open-source launch checklist and operating system for solo SaaS builders.

Make Me Unicorn

Catch what AI-generated SaaS code misses — before it costs you users, money, or trust.

The open-source launch checklist and operating system for solo builders.

English · 🇰🇷 한국어 · 🇯🇵 日本語 · 🇨🇳 简体中文 · 🇪🇸 Español

60-Second Demo

pip install make-me-unicorn
cd your-project
mmu init && mmu scan && mmu

  MAKE ME UNICORN - STATUS DASHBOARD

          .--*--.
         / *v*  \
        |       |
         \ ___ /
          '---'

  Stage: HATCHING    ######..............  22%  (124/551)

  LAUNCH GATES  (16/26)
    M0 Problem Fit         ################   4/4   PASS
    M1 Build Fit           ################   5/5   PASS
    M2 Revenue Fit         ############....   3/4   OPEN
    M3 Trust Fit           ################   4/4   PASS
    M4 Growth Fit          ########........   2/4   OPEN
    M5 Scale Fit           ####............   1/5   OPEN

Your unicorn evolves as you build: Egg → Hatching → Foal → Young → Unicorn → Legendary.

Add a Badge to Your README

Show your launch readiness to the world. One command:

mmu badge                          # get markdown badge
mmu badge --format svg -o badge.svg  # save as SVG file
mmu badge --clipboard              # copy to clipboard

Then paste in your README:

Every badge links back here. Every README becomes a growth channel.

Share Your Score

mmu share                     # print shareable score card
mmu share --clipboard         # copy to clipboard (macOS)

┌─────────────────────────────────────────────┐
│  Make Me Unicorn — Launch Readiness         │
│                                             │
│  Score: 68%  Stage: YOUNG UNICORN           │
│                                             │
│  M0 Problem Fit    ████████████████  PASS   │
│  M1 Build Fit      ████████████████  PASS   │
│  M2 Revenue Fit    ██████████░░░░░░  OPEN   │
│  M3 Trust Fit      ████████████████  PASS   │
│  M4 Growth Fit     ████████░░░░░░░░  OPEN   │
│  M5 Scale Fit      ████░░░░░░░░░░░░  OPEN   │
│                                             │
│  Stack: Next.js · Stripe · SSR              │
│  pip install make-me-unicorn                │
│  github.com/minjikim89/make-me-unicorn      │
│  #MakeMeUnicorn                             │
└─────────────────────────────────────────────┘

Check your SaaS launch readiness:
https://github.com/minjikim89/make-me-unicorn

The Problem

You code with AI. You ship fast. But then:

What goes wrong	What it costs you
Forget password reset while building login	Users locked out on day 1
Skip webhook signature verification	Attackers replay payment events
Launch without OG tags	Every shared link looks broken
Lose context between AI sessions	Re-explain your project from scratch
No refund policy	First dispute = frozen Stripe account

You're not failing at coding. You're failing at tracking what matters.

What MMU Covers

Building Frontend (responsive, a11y, forms) Backend (API, DB, queues) Auth (login, reset, OAuth, sessions) Billing (Stripe, webhooks, refunds) Testing (unit, E2E, agent safety)

Launching SEO (OG tags, sitemap, meta) Legal (privacy, terms, GDPR) Security (CORS, rate limits, secrets) Performance (caching, lazy load) CI/CD (pipeline, rollback plan)

Operating Monitoring (errors, uptime, alerts) Analytics (funnel, retention, events) Email (transactional, templates) Accessibility (WCAG, keyboard nav) Reliability (backup, incident plan)

534+ items. 15 categories. 6 launch gates. Zero guesswork.

Personalize Your Checklist

Not every project needs billing or i18n. MMU adapts:

mmu init                      # generates .mmu/config.toml

[features]
billing = false               # no Stripe? billing items won't count against you
i18n = false

[architecture]
framework = "nextjs"

Your score reflects only what applies to your project. mmu status --why shows the breakdown — like Lighthouse, but for SaaS readiness.

6 Launch Gates

Phase exits. Don't skip ahead.

M0 Problem Fit    →  Do you know WHO and WHY?
M1 Build Fit      →  Does the core product work end-to-end?
M2 Revenue Fit    →  Can someone pay you? And get a refund?
M3 Trust Fit      →  Privacy policy? Support path? Logging?
M4 Growth Fit     →  Will links look right? Can people find you?
M5 Scale Fit      →  What happens at 3am?

Run mmu gate --stage M0 to verify.

12 Operating Modes

One mode per session. Each loads only the docs you need — prevents the #1 problem with AI coding: context overload.

mmu start --mode backend      # loads: architecture.md, sprint, ADR logs
mmu start --mode billing      # loads: pricing.md, billing checklist, compliance
mmu start --mode growth       # loads: SEO checklist, metrics

AI Integration (Optional)

MMU works without AI. But with Claude, it gets powerful:

pip install make-me-unicorn[llm]
export ANTHROPIC_API_KEY=sk-ant-...

Command	What happens
mmu init --interactive	Answer 5 questions → Claude writes your strategy, product, pricing, architecture, and UX docs
mmu start --mode X --agent	Auto-formats session context for Claude Code or any LLM
mmu doctor --deep	Claude reads your code, flags mismatches, security gaps, blind spots
mmu generate strategy	Generates or updates core docs based on current project state

Core CLI stays zero-dependency. AI features degrade gracefully.

Full Command Reference

mmu                           # status dashboard
mmu init                      # bootstrap project
mmu init --interactive        # LLM-guided setup (5 questions → 5 docs)
mmu scan                      # auto-detect tech stack
mmu status --why              # score breakdown
mmu next                      # prioritized next actions
mmu show frontend             # drill into any category
mmu check frontend 3          # mark item done
mmu gate --stage M0           # verify gate readiness
mmu doctor                    # guardrail health checks
mmu doctor --deep             # LLM-powered semantic review
mmu share                     # shareable score card
mmu badge                     # README badge (markdown/svg/html)
mmu start --mode backend      # start focused session
mmu close                     # end session with structured memory

Who This Is For

You are...	MMU helps you...
A founder coding with AI	Stop re-explaining your project. Keep context across tools.
A frontend developer	Know exactly what to build: auth flows, error states, OG tags.
A product manager	Structured PRD, pricing strategy, launch checklist — all in markdown.
A fullstack builder	Track everything in one place. Nothing slips through.

Example: TaskNote

A fully filled-out example:

examples/filled/tasknote/
├── docs/core/strategy.md      ← ICP, value prop, competitors
├── docs/core/product.md       ← MVP scope, user journey, P0/P1
├── docs/core/pricing.md       ← Free/Pro/Team, billing rules
├── docs/core/architecture.md  ← Next.js + FastAPI + Postgres
├── docs/adr/001_billing_provider_choice.md  ← Why Stripe?
└── current_sprint.md          ← This week's 3 goals

Requirements

• Python 3.10+
• Core CLI: zero external dependencies
• AI features: pip install make-me-unicorn[llm]

CI Guardrails

mmu doctor runs on every PR. mmu gate runs for stages listed in docs/ops/gate_targets.txt.

Contributing

See CONTRIBUTING.md.

License

MIT. See LICENSE.