agentic_kg_memory

What problem does it solve?

It helps you turn scattered source material into graph-backed semantic memory, retrieve the most relevant evidence for a question, and continuously update conclusions as new evidence arrives.

Core Features & Use Cases

• Graph-backed semantic memory: extract normalized triplets (with polarity and inference type) and store them as typed edges plus searchable dense/sparse surfaces.
• Evidence-first retrieval and synthesis: narrow candidates with BM25/vector similarity, optionally route structural queries via the KG, and synthesize answers grounded in sources.
• Updateable “throughlines”: maintain durable abductive conclusions whose identity is the premise set, then revise or supersede them as evidence changes.

Quick Start

Run agentic_kg_memory to ingest your sources, build the triplet and page layers, and retrieve grounded throughlines for your next question.