• Doc Changes

Added #

• Strongly typed generic NeuroController<T>
• Generic NeuroPredictor<T> interface
• Typed state model support (no more dynamic Map-based state)
• Built-in loading state handling (isLoading)
• Built-in error handling (error)
• preview() method for non-committed predictive state evaluation
• setState() method for controlled manual state updates
• Improved example application with typed architecture

Changed #

• Replaced dynamic Map<String, dynamic> state with strongly typed state model
• Improved controller architecture for better scalability
• Enhanced internal state safety and predictability
• Updated README with full typed usage documentation

Improved #

• Cleaner API design
• Better separation of concerns between controller and predictor
• More production-ready structure
• Improved extensibility for future AI/LLM integrations

Notes #

This release transitions NeuroState from an experimental prototype to a production-ready, strongly-typed predictive state management architecture for Flutter.

Added #

• Core NeuroController for predictive state management
• NeuroPredictor interface for pluggable AI logic
• RuleNeuroPredictor for offline intelligent predictions
• Event-driven predictive state dispatching
• Example Flutter demo showcasing NeuroState usage

Notes #

Initial experimental release introducing AI-augmented state management concept for Flutter apps.