Knowledge

You can build a personal knowledge graph by storing conversations and documents as structured episodes. The system automatically extracts entities and maps relationships, enabling persistent memory and semantic search retrieval using natural language queries.

Yes, this knowledge graph system supports both Neo4j and FalkorDB backends. It automates entity extraction and relationship mapping from your inputs, storing them as structured episodes for persistent retrieval and graph traversal.

Semantic search in this knowledge graph allows you to find information using natural language queries. It retrieves relevant entities, facts, and relationships by traversing the graph structure and matching the semantic meaning of your question.

The best way to store conversation context is by saving inputs as structured episodes in your knowledge graph. The system captures temporal context automatically, preserving the time-sensitive nature of your data for accurate historical retrieval.

Yes, the knowledge graph supports OSINT and CTI ontology for investigative search. You can explore how different pieces of knowledge are related through graph traversal, making it ideal for threat hunting and understanding complex topics.

Memory decay lifecycle management is supported natively by the knowledge graph. This allows you to automatically degrade or retire outdated information over time, ensuring your persistent memory remains relevant and accurate for semantic search.