Embed GraphRAG-style retrieval workflows within an embedded DuckDB store.
npx skills add https://github.com/bradAGI/GraphMemory --skill graphmemory
Please help me install this Agent Skill. Skill: graphmemory Source: https://github.com/bradAGI/GraphMemory/tree/main Command: npx skills add https://github.com/bradAGI/GraphMemory --skill graphmemory
GraphMemory provides an embedded GraphRAG knowledge graph platform built on DuckDB that combines vector, full-text, and hybrid search to store, retrieve, and reason over entities and relationships.
Install graphmemory and create a GraphMemory instance, add a few Node/Edge objects, and run a simple retrieve or search.
High-intent search queries and answers about installing and using this skill.
Build a GraphRAG knowledge graph by embedding nodes and edges in an in-memory or disk-backed DuckDB store, then use hybrid search, graph expansion, and multi-hop traversal to retrieve context for LLM-guided Q&A.
Extract entities and relationships from unstructured text for a knowledge graph using optional DSPy-based extraction, then merge or upsert nodes and edges with fuzzy deduplication to maintain graph consistency.
Perform hybrid search in a DuckDB graph by blending BM25 full-text search across node properties with HNSW vector similarity using cosine, L2, or inner product metrics to return more relevant retrieved nodes.
GraphMemory supports importing and exporting graph data in JSON, CSV, and GraphML formats, and includes a browser-based visualizer to inspect the embedded knowledge graph structures.
Handle duplicate nodes and edges during upserts into a vector database knowledge graph using built-in fuzzy deduplication and upsert semantics, ensuring transactional safety with automatic retries for transient errors.
Use NetworkX graph algorithms like PageRank and centrality in an embedded GraphRAG pipeline when you need to analyze component structures and compute node importance to refine vector and text search context.
seb1nConstructs structured knowledge graphs from unstructured text with Cypher, JSON-LD, and Mermaid outputs.
mkurmanCreate, analyze, and visualize complex networks and graphs in Python.
sparq-orgAnalyze RDF graphs with PageRank, degree centrality, and label propagation.
ricableRun embedded vector searches with SQL, SPARQL, and Cypher via WASM.
