A multifractal AI system with real-time search, teachable responses, and open-source modularity
Version files:
- Alexandrian i6-v2.9.5 (optimized) Filename: server_hexpre295.py
- Alexandrian i6-v2.9.6 (slower) more comprehensive response generation. Filename: server_hexpre296.py
- Fetches real-time data from Wikipedia, DictionaryAPI, and DuckDuckGo
- Hybrid response generation combining multiple authoritative sources
- Context-aware sentence filtering using NLP-powered relevance scoring
self.resources = [ ("https://api.newservice.com/{}", self._parse_custom_handler), # Add your custom API endpoint and parser ]
🎓 Teachable Interface
Teach Mode: Use Respond: [your answer] to train custom responses
Persistent memory via inputs.json for personalized knowledge retention
Feedback-driven learning with automatic preference adjustment
🛠 Open Source Modularity
Fully customizable NLP pipeline (NeuroLinguisticProcessor class)
Extensible hexagram-based decision core (HexagramGrid system)
Clear documentation for modifying:
Response logic (ResponseLogic*.py)
API handlers (OmniSourceNLP*.py)
Multifractal analysis algorithms
🚀 Getting Started Installation
pip install requests beautifulsoup4 colorama numpy matplotlib aiohttp
Launch
python server_hexpre295.py
🧠 Usage Examples
Basic Query [User]: Explain quantum entanglement
Teach Mode [User]: Respond: Quantum entanglement is when particles share states instantaneously across distance
API Integration
resources.append(("https://api.academic.com/{}", _parse_academic_db))
System Commands
help Show command list print multifractal spectrum Display cognitive analysis spectrum plot multifractal spectrum Visualize decision-making patterns plot multifractal spectrum log Visualize decision-making patterns 🧩 Customization Guide Add New Knowledge Sources
Create handler function in OmniSourceNLP2.py
Add endpoint to self.resources array
Implement parsing logic for API responses
Modify Response Logic
Adjust relevance thresholds in ResponseParameters class
Tune multifractal weights in NeuroMorphicProcessor
Customize stop words in _load_stop_words()
📜 Philosophy
"An open mind that grows with its users" Alexandrian combines:
Adaptive Learning: Stores user corrections in JSON
Fractal Cognition: Uses I Ching-inspired decision matrices
Transparent AI: All decision paths visible in code
🤝 Contributing
We welcome PRs for:
New API integrations
Enhanced NLP rules
Improved error handling
Additional documentation
License: Alexandrian i6 Universal License
📘 Full Documentation | 🐛 Issue Tracker | 💡 Feature Requests
"Wisdom is not a destination, but a continually evolving path." - Alexandria System Prompt v2.9.5