Neo4j Graphrag Skill

If you're wiring up retrieval-augmented generation on Neo4j and need to pull graph context alongside your vectors, this is the skill to reach for. It covers the neo4j-graphrag Python package's full retriever lineup: plain vector, hybrid with fulltext, and the Cypher variants that let you traverse relationships after the initial embedding lookup. The decision tree is clean: use HybridCypherRetriever for production GraphRAG, VectorRetriever for a quick baseline, Text2CypherRetriever when you want natural language to Cypher, and ToolsRetriever when the LLM should route between multiple strategies. It also handles LLM provider setup across OpenAI, Anthropic, Vertex, Bedrock, and others, plus the newer SEARCH clause mechanics in Cypher 25. Doesn't touch knowledge graph construction or GDS analytics; those live in separate skills.

skills.sh installs: 1.2K
GitHub stars: 121
last updated: Aug 21, 2026

View on GitHub

Install

$ npx -y skills add neo4j-contrib/neo4j-skills --skill neo4j-graphrag-skill --agent claude-code

Installs into .claude/skills of the current project.

Files

Neo4j GraphRAG Skill

When to Use

Building GraphRAG retrieval pipelines with neo4j-graphrag Python package
Choosing between VectorRetriever, HybridRetriever, VectorCypherRetriever, HybridCypherRetriever
Writing retrieval_query Cypher fragments for graph-augmented context
Wiring retriever + LLM into a GraphRAG pipeline
Using LLM-routed multi-retriever with ToolsRetriever
Debugging low retrieval quality
Integrating Neo4j with LangChain, LlamaIndex, or Haystack

When NOT to Use

KG construction from documents → neo4j-document-import-skill
Plain vector/semantic search without graph traversal → neo4j-vector-index-skill
Hybrid search that combines vector with fulltext or other ranked sources → neo4j-vector-index-skill
GDS algorithms (PageRank, Louvain, node embeddings) → neo4j-gds-skill
Agent long-term memory → neo4j-agent-memory-skill
Writing raw Cypher queries → neo4j-cypher-skill

Retriever Selection

Has fulltext index?
  YES → Hybrid variants (HybridRetriever / HybridCypherRetriever)
  NO  → Vector variants (VectorRetriever / VectorCypherRetriever)

SKILL.md on GitHub