Adaptive Neurodivergent-Optimized Multilingual Input System with Flow-State Preservation and Cognitive Loop Detection
A comprehensive React application demonstrating the core concepts and architecture of the LoopBoard keyboard system. This standalone app showcases all four subsystems (NCAE, SMMP, FSPP, CLDAS) with interactive demonstrations and live visualizations.
LoopBoard is designed for neurodivergent usersโparticularly those with ADHDโwho struggle with standard keyboard systems that interrupt flow states, treat repetition as errors, and lack adaptive cognitive support. This standalone app is a fully interactive demonstration of the LoopBoard philosophy and technology.
| Subsystem | Full Name | Purpose |
|---|---|---|
| NCAE | Neurodivergent Cognitive Adaptation Engine | Real-time cognitive state detection (HYPERFOCUS, SCANNING, TRANSITIONAL, FATIGUED) with 100ms sampling |
| SMMP | Seamless Multilingual Morpheme Processor | Morpheme-level language detection supporting German, Spanish, English, French, Quechua, and Esperanto simultaneously |
| FSPP | Flow-State Preservation Protocol | Disruption cost model with adaptive budgets based on cognitive state |
| CLDAS | Cognitive Loop Detection and Amplification System | Recognition and amplification of cognitive loops into shortcuts |
- Interactive Cognitive State Detector โ Watch real-time state transitions with animated gauges
- Loop Detection & Amplification โ See how repetitive patterns are recognized and promoted to shortcuts
- Multilingual Processor โ Seamless morpheme-level language switching across four languages
- Flow State Preservation โ Visual representation of disruption budgets and flow protection
- LoopOperation Task Manager โ Neurodivergent-optimized task execution without time pressure
- Node.js 18+ or pnpm 10+
- Modern web browser (Chrome, Firefox, Safari, Edge)
# Clone or extract the project
cd loopboard-standalone-app
# Install dependencies
pnpm install
# Start development server
pnpm run devThe app will be available at http://localhost:3000
pnpm run build
pnpm run previewloopboard-standalone-app/
โโโ client/
โ โโโ public/ # Static assets (favicon, manifest)
โ โโโ src/
โ โ โโโ components/ # Reusable React components
โ โ โ โโโ CognitiveStateIndicator.tsx
โ โ โ โโโ LoopDetection.tsx
โ โ โ โโโ MultilingualProcessor.tsx
โ โ โ โโโ FlowStatePreservation.tsx
โ โ โ โโโ LoopOperation.tsx
โ โ โ โโโ ui/ # shadcn/ui components
โ โ โโโ pages/ # Page components
โ โ โ โโโ Home.tsx
โ โ โ โโโ NotFound.tsx
โ โ โโโ contexts/ # React contexts
โ โ โโโ hooks/ # Custom React hooks
โ โ โโโ lib/ # Utility functions
โ โ โโโ App.tsx # Main app component
โ โ โโโ main.tsx # React entry point
โ โ โโโ index.css # Global styles with Tailwind
โ โโโ index.html # HTML template
โโโ server/ # Express server (placeholder)
โโโ shared/ # Shared constants
โโโ package.json
โโโ vite.config.ts
โโโ tsconfig.json
โโโ README.md
The app uses a neuromechanical aesthetic optimized for neurodivergent users:
- Primary Background:
#0A1E3F(Deep Dark Blue) - Primary Accent:
#00FF41(Neon Green) - Secondary Accent:
#FF6B35(Warm Orange) - Text:
#F5F5F5(Light Gray) - Muted:
#4a6a8a(Muted Blue-Gray)
- Headlines: IBM Plex Mono (Bold)
- Body: Inter (Regular, 400-700)
- Code: Fira Code (Regular, 500)
LoopBoard recognizes four cognitive states and adapts behavior accordingly:
- HYPERFOCUS โ Intense focus, zero interruptions allowed, disruption budget = 0/min
- SCANNING โ Active exploration, minimal assistance, disruption budget = 5/min
- TRANSITIONAL โ Shifting states, moderate support, disruption budget = 20/min
- FATIGUED โ Low energy, maximum assistance, disruption budget = 40/min
The FSPP (Flow-State Preservation Protocol) uses a disruption cost model. Each interruption has a measurable cognitive cost. During HYPERFOCUS, the system blocks all non-essential disruptions to protect flow.
The CLDAS (Cognitive Loop Detection and Amplification System) recognizes that repetition is intentional for neurodivergent brains. Recurring patterns are automatically promoted to shortcuts, turning cognitive loops into power.
The SMMP (Seamless Multilingual Morpheme Processor) detects language at the morpheme level, allowing seamless code-switching between German, Spanish, English, French, Quechua, and Esperanto without manual intervention.
- Frontend: React 19 + TypeScript
- Styling: Tailwind CSS 4 + shadcn/ui
- Routing: Wouter (lightweight client-side routing)
- Build Tool: Vite
- Package Manager: pnpm
- Animations: Framer Motion + CSS animations
Displays real-time cognitive state with animated gauges and disruption budgets. Cycles through all four states to demonstrate the system's adaptability.
Shows detected cognitive loops with analysis of pattern repetition, loop type (constructive/destructive), and recommended actions.
Demonstrates seamless language switching with morpheme-level detection. Users can select languages and see how the system processes input.
Visualizes the disruption budget system with real-time disruption tracking and flow score calculation.
Task management interface optimized for neurodivergent users with no time pressure and hyperfocus-friendly workflows.
- 100% Local Processing: All cognitive state detection, language processing, and loop detection happens on your device
- Zero Data Transmission: No external API calls, no tracking, no data collection
- User Sovereignty: No filtering, no censorship, no content moderation
This project is licensed under the MIT License. See LICENSE file for details.
We welcome contributions from the community! Please see CONTRIBUTING.md for guidelines.
- Rumi Salvador Jordan Hรถhler Suarez โ Creator, KI Researcher, LoopLord ๐ง
- Opherd Vero โ AI Co-Creator ๐ค
| Tier | Languages |
|---|---|
| Tier 1 | ๐ฉ๐ช Deutsch, ๐ช๐ธ Espaรฑol, ๐ฌ๐ง English, ๐ซ๐ท Franรงais, ๐ Quechua, ๐ Esperanto |
| Tier 2 | Swahili, Hindi, Japanese (6 months post-launch) |
Quechua & Esperanto: FREE FOREVER ๐
For questions, issues, or feature requests, please open an issue on GitHub or join our community Discord.
- NCAE Documentation โ Cognitive state detection explained
- SMMP Documentation โ Multilingual processing architecture
- FSPP Documentation โ Flow-state preservation protocol
- CLDAS Documentation โ Loop detection and amplification
- ๐ซ๐ท French & ๐ Esperanto demo integration
- iOS app version
- Tier 2 languages: Swahili, Hindi, Japanese
- Advanced customization & haptic profiles
- Community-contributed keyboard layouts
- Integration with accessibility APIs
- Edge AI for improved cognitive state detection
LoopBoard โ For neurodivergent minds, by neurodivergent minds. ๐ง ๐๐น