Official website: https://www.sbti.ai/en

SBTI is an internet-native personality experience focused on entertainment: around 30 questions (plus a hidden branch trigger), cross-analysis across five core models and fifteen dimensions, and shareable personality outcomes.

This repository is the open-source codebase for the SBTI web app and integration architecture.

Disclaimer: For entertainment and self-reflection only—not scientific diagnosis, education, or hiring. This project is an original parody with its own identity; it is not affiliated with any similarly named organization or website.

Screenshots: add PNGs under docs/screenshots/ (see docs/screenshots/README.md) and embed them here for GitHub.

• Landing (/) — hero, example type cards, integrations overview
• Main test (/test) — animated flow, progress, Zustand + local persistence
• Results (/result/[type]) — full type card, compatibility links, share payload, OG image
• Compare (/compare) — overlap/contradiction, “appearance vs reality” blurb, radar chart
• Integrations (/integrations, /integrations/[provider]) — registry of adapters
• Dev JSON lab (/dev/pack) — validate custom trait JSON
• API — POST /api/compare, GET /api/share?type=EOVA, GET /api/session
• Persistence — Prisma + SQLite (dev); swap DATABASE_URL for Postgres in production

• Core test shape — ~31-question flow, includes hidden-branch style logic, optimized for 5–8 minutes
• Five core models — Self, Emotional, Attitude, Action Drive, Social
• 15-dimension mindset — model-backed scoring and mapping into personality outputs
• Cross-analysis direction — supports external personality systems and MBTI-style adapter mapping
• Privacy-first messaging — browser-local calculation style can be supported by client-only mode

Four axes are derived from weighted answers (E/I, O/L, V/M, A/P), each on a 0–100 scale. The four letters are chosen by whether each score is ≥ 50. The 16 type codes are the Cartesian product of those dichotomies; copy and metadata live in src/lib/sbti/types-catalog.ts.

Each adapter implements PersonalityTestAdapter:

• normalizeResult — maps manual / quiz / JSON input → NormalizedTraitProfile
• mapToSBTI — uses mapNormalizedToSBTI (shared rules engine)
• getResultLabel — human-readable summary for history rows

Built-in examples: Four-Letter Compass (style), Ocean Five (style), Nine Archetype Map (style), Custom JSON trait pack.

1. Copy src/tests/registry/example-provider.template.ts.
2. Implement normalizeResult + getResultLabel; reuse mapNormalizedToSBTI for mapping.
3. Call registerPersonalityProvider(yourAdapter) from a small module imported by src/lib/providers/registry.ts (or a dedicated register-extra.ts).
4. Run npm test and exercise /compare + /dev/pack.

npm install
cp .env.example .env
npx prisma migrate dev
npm run db:seed      # optional demo user + sample comparisons
npm run dev
npm test
npm run build

1. Push this repo to GitHub and import into Vercel.
2. Set DATABASE_URL to a hosted Postgres URL (recommended for serverless). Run npx prisma migrate deploy in the build command or as a release step.
3. Set NEXT_PUBLIC_SITE_URL to your production URL (Open Graph + metadata base).
4. SQLite file URLs are fine for local dev only—not ideal on serverless volumes.

Next.js (App Router) · TypeScript · Tailwind CSS · shadcn/ui (Base UI) · Framer Motion · Zustand · Zod · Prisma 6 · Vitest · Recharts

Optional: Auth.js is listed in dependencies for future email/OAuth; the app ships with anonymous cookie sessions via /api/session.

Implemented in src/lib/functions.ts:

takeSBTITest, scoreSBTIResult, generateSBTIType, saveSubmission, importExternalPersonalityResult, normalizeExternalTestResult, comparePersonalitySystems, registerPersonalityProvider, getAvailablePersonalityProviders, compareWithProvider, generateShareCardPayload, exportUserProfile

• Optional Auth.js flows + account linking
• Server-side share PNG generation (beyond JSON download)
• More first-party “mini quizzes” per adapter
• Plugin hot-loading from /src/tests/registry without editing core registry

See CONTRIBUTING.md. Licensed under MIT.