Jnana (Sanskrit: ज्ञान, meaning "knowledge" or "wisdom") is an integrated AI co-scientist platform that combines the interactive capabilities of Wisteria with the scalable multi-agent system of ProtoGnosis.

Jnana provides researchers with:

• Interactive Hypothesis Generation: Real-time hypothesis creation and refinement
• Multi-Agent Processing: Scalable automated hypothesis evaluation and ranking
• Scientific Rigor: Comprehensive evaluation against established scientific criteria
• Flexible Deployment: Support for interactive, batch, and hybrid workflows

Jnana integrates two powerful systems:

• Wisteria: Interactive terminal-based hypothesis generation with real-time feedback
• ProtoGnosis: Multi-agent system for scalable research hypothesis processing

• Professional curses-based multi-pane interface
• Real-time hypothesis refinement with user feedback
• Scientific hallmarks evaluation (testability, specificity, etc.)
• Professional PDF export with comprehensive documentation
• Session management and version tracking

• Multi-agent system with specialized roles
• Tournament-based hypothesis ranking
• Support for multiple LLM providers
• Asynchronous processing for large datasets
• Comprehensive research workflow automation

• Seamless switching between interactive and automated modes
• Unified data model supporting both systems
• Combined evaluation framework
• Enhanced model management
• Comprehensive research session documentation

# Clone the repository
git clone <repository-url>
cd Jnana

# Install dependencies
pip install -r requirements.txt

# Configure models
cp config/models.example.yaml config/models.yaml
# Edit config/models.yaml with your API keys and model preferences

python jnana.py --mode interactive --goal "Your research question here"

python jnana.py --mode batch --goal "Your research question here" --count 20

python jnana.py --mode hybrid --goal "Your research question here" --interactive-refinement

Jnana/
├── jnana/                    # Core Jnana package
│   ├── core/                 # Core integration components
│   ├── data/                 # Unified data models
│   ├── ui/                   # User interface components
│   ├── agents/               # Agent wrappers and extensions
│   └── utils/                # Utility functions
├── config/                   # Configuration files
├── tests/                    # Test suite
├── docs/                     # Documentation
├── examples/                 # Example usage scripts
└── scripts/                  # Utility scripts

This project is currently in Phase 1 development, focusing on:

• Repository structure setup
• Unified data model implementation
• Model configuration integration
• Core integration framework
• Basic UI integration

Please see CONTRIBUTING.md for development guidelines.

[License information to be added]

• Wisteria: Interactive hypothesis generation system by Rick Stevens AI
• ProtoGnosis: Multi-agent AI co-scientist system