Version: 1.0.0
Last Updated: 2026-02-04
The AI reformer. Knows what it doesn't. Teams with you to get it right.
The AI and the human work as a team to get results through structured aligned discussions.
The prompt is part of a structured aligned discussion called hum&ai, where the AI and the human work as a team to get results.
The discussion runs in threads (lines of thought), with strolls (a sequence of request-response iterations on a topic) possibly broken into sections (a subtopic stroll) and if needed, further broken into parts.
This way, without losing information, long detailed responses are cut up into short parts, each discussed in several aligned iterations, proceeding at a human pace.
The discussion produces (or updates) a brief and concise HOCON report to support the hum&ai process.
Planned threads: The remainder of topics not discussed yet, in this and other threads is remembered and maintained with alignment discussions.
See devDocs/TechSetup.md for the full directory layout and tech stack details.
- React Native - Mobile application framework
- Expo - React Native development platform
- React Navigation - Navigation library
- Node.js - JavaScript runtime
- NATS.io - Message broker for microservices communication
- LangChain - Framework for developing LLM applications
- LangGraph - Library for building stateful, multi-actor applications
- Node.js >= 18.0.0
- npm >= 9.0.0
- (Optional) NATS server for local development
- (Optional) Expo Go app on your mobile device
git clone https://github.com/pashute/BinAI.ai.git
cd BinAI.ainpm installThis will install dependencies for the root, frontend, and backend workspaces.
cd backend
cp .env.example .envEdit .env and add your API keys:
OPENAI_API_KEY- For LangChain/OpenAI integrationNATS_SERVER- NATS server URL (default: nats://localhost:4222)
From the root directory:
npm run devThis will start both the humandBrain backend and humANDai frontend concurrently:
- humandBrain: Runs on port 4222 (NATS) / 3000 (default)
- humANDai: Expo dev server (Metro bundler)
# Start only backend
npm run dev:humandBrain
# Start only frontend
npm run dev:humANDai
# Clean all dependencies and build artifacts
npm run cleanThis project is configured for GitHub Codespaces with VSCode:
- Open the repository in GitHub Codespaces
- The devcontainer will automatically set up the environment
- English locale and Windows-compatible terminal settings are pre-configured
- GitHub Copilot extensions are included
The following ports are automatically forwarded:
8081- Expo Metro bundler3000- Backend API (if used)4222- NATS server
- Install Expo Go app on your device
- Start the dev server:
npm run dev - Scan the QR code shown in terminal with:
- iOS: Camera app
- Android: Expo Go app
.devcontainer/devcontainer.json- Codespace configuration.devcontainer/Dockerfile- Container definition- English locale and Windows terminal preferences included
frontend/app.json- Expo app configurationfrontend/babel.config.js- Babel transpiler configuration
humandBrain/.env- Environment variables (create from .env.example)
npm run dev- Run both frontend and backend concurrentlynpm run dev:humandBrain- Run only backendnpm run dev:humANDai- Run only frontendnpm run install:all- Install all dependenciesnpm run clean- Clean all workspaces
npm run dev --workspace=humandBrain- Run backend with hot reloadnpm run start --workspace=humandBrain- Run backend in production mode
npm run start --workspace=humANDai- Start Expo dev servernpm run android --workspace=humANDai- Run on Android emulatornpm run ios --workspace=humANDai- Run on iOS simulatornpm run web --workspace=humANDai- Run as web app
See humandBrain/.env.example for all available environment variables.
Required for full functionality:
OPENAI_API_KEY- OpenAI API key for LangChainNATS_SERVER- NATS messaging server URL
Details will be provided after initial requirements gathering.
Details will be provided.
More features and requirements will be added as the project evolves.
See DevDocs/Frontend/features.app.md and DevDocs/Backend/features.brain.md for current feature plans.
Note: This is an initial setup. Detailed Requirements and features will be implemented in subsequent iterations.