skills.rest
eklyukin
eklyukinCOMMUNITY
Explore all skills by this author →
GitHub

neo4j-graphrag-skill

Build GraphRAG retrieval pipelines on Neo4j using the neo4j-graphrag Python package.

Updated Jul 24, 2026
Search Tags:#llm#retrieval#vector search#rag pipeline#neo4j#cypher#graphrag
One-click install
npx skills add https://github.com/eklyukin/my-ai-config --skill neo4j-graphrag-skill-eklyukin
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: neo4j-graphrag-skill
Source: https://github.com/eklyukin/my-ai-config/tree/main/skills/neo4j-graphrag-skill
Command: npx skills add https://github.com/eklyukin/my-ai-config --skill neo4j-graphrag-skill-eklyukin

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill requires neo4j-graphrag, neo4j, and includes references (resource) components.

What problem does it solve? Choosing and wiring the right retriever, embedder, and LLM for a Neo4j-backed GraphRAG pipeline involves many version-specific APIs and failure modes (index state, retrieval_query syntax, score injection). This Skill provides tested patterns and decision tables so you get a working retrieval pipeline without trial and error. ## Core Features & Use Cases - Retriever Selection: Decision tables for VectorRetriever, HybridRetriever, VectorCypherRetriever, HybridCypherRetriever, Text2CypherRetriever, ToolsRetriever, and external vector DB retrievers (Weaviate, Pinecone, Qdrant). - Pipeline Wiring: Complete GraphRAG patterns covering retrieval_query Cypher fragments, query_params, filters, custom prompt templates, message history, and token usage tracking across OpenAI, Anthropic, VertexAI, Bedrock, Cohere, Mistral, and Ollama. - Use Case: You have Chunk nodes with embeddings in Neo4j and want to answer questions like "Who does Alice work for?" by combining vector similarity with graph traversal over related Article and Organization nodes, then feeding the context to an LLM. ## Quick Start Use the neo4j-graphrag skill to build a HybridCypherRetriever pipeline that retrieves chunks from my Neo4j database and answers questions with an LLM.

Frequently Asked Questions about neo4j-graphrag-skill

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I build a GraphRAG pipeline with Neo4j in Python?▼

Install neo4j-graphrag, create vector and fulltext indexes, then wire a retriever (such as HybridCypherRetriever) with an embedder and LLM into a GraphRAG object. Call rag.search() with your query text and retriever_config to get grounded answers.

Which neo4j-graphrag retriever should I use?▼

Use HybridCypherRetriever as the production default when you have both vector and fulltext indexes and need graph traversal. Use VectorRetriever for baseline semantic search, Text2CypherRetriever for natural-language-to-Cypher without an embedder, and ToolsRetriever for LLM-routed multi-retriever setups.

Does neo4j-graphrag support Anthropic or local LLMs?▼

Yes, neo4j-graphrag supports AnthropicLLM, OpenAILLM, VertexAILLM, BedrockLLM, CohereLLM, MistralAILLM, and OllamaLLM for local models. All implement LLMBase with sync and async invocation, tool calling, and automatic rate limiting.

Why does my retrieval_query return zero rows or a KeyError on score?▼

Zero rows usually means a wrong relationship direction or missing MATCH in the fragment; verify with EXPLAIN and db.schema.visualization(). A KeyError on score means the RETURN clause omits score, which is auto-injected and must appear in every retrieval_query RETURN.

Can I use vectors stored in Pinecone or Weaviate with Neo4j GraphRAG?▼

Yes, WeaviateNeo4jRetriever, PineconeNeo4jRetriever, and QdrantNeo4jRetriever map external vector store IDs to Neo4j node IDs and run a retrieval_query for graph context. Install the matching extra such as neo4j-graphrag[pinecone].

When should I not use the neo4j-graphrag retrievers?▼

Use neo4j-document-import-skill for KG construction from documents, neo4j-vector-index-skill for plain vector or hybrid search without graph traversal, neo4j-gds-skill for graph algorithms, and neo4j-agent-memory-skill for agent long-term memory.

Related Skills

Explore More Software Engineering Skills→
neo4j-contribneo4j-contrib
101

neo4j-document-import-skill

Ingest PDFs and text into Neo4j knowledge graphs via SimpleKGPipeline.

Official
Advanced
Clay-HHKClay-HHK
1

langchain

Build LLM applications with agents, chains, RAG pipelines, and vector store retrieval.

Community
Intermediate
kevinsweetkevinsweet

rag-chatbot

Build RAG chatbots that ground answers in data using Elasticsearch retrieval.

Community
Advanced
neo4j-contribneo4j-contrib
101

neo4j-graphrag-skill

Build GraphRAG retrieval pipelines on Neo4j with neo4j-graphrag retrievers.

Official
Advanced