Graphrag Patterns

This one teaches you to build RAG systems that understand relationships between entities, not just semantic similarity. You'll extract entities and connections from documents, store them in Neo4j, then retrieve context by traversing the graph. It covers three retrieval patterns: entity-centric (expand from query entities to neighbors), path-based (find connections between entities), and community-based (the Microsoft GraphRAG approach with hierarchical clustering). The integration examples with LangChain are solid, including natural language to Cypher queries. Use this when your data has meaningful connections and you need multi-hop reasoning, like "How is X related to Y through Z?" Standard vector search can't do that.

skills.sh installs: 207
GitHub stars: 6
last updated: Aug 21, 2026

View on GitHub

Install

$ npx -y skills add latestaiagents/agent-skills --skill graphrag-patterns --agent claude-code

Installs into .claude/skills of the current project.

Files

GraphRAG Patterns

Combine knowledge graphs with RAG for complex reasoning over connected data.

When to Use GraphRAG vs Vector RAG

Use Case	Vector RAG	GraphRAG
Simple Q&A	✅	Overkill
Factual lookup	✅	✅
Multi-hop reasoning	❌	✅
"How is X related to Y?"	❌	✅
Entity relationships	❌	✅
Compliance/audit trails	❌	✅
Summarizing themes	❌	✅
Explaining why items are connected	❌	✅ (the path is the explanation)