neo4j-getting-started-skill

To build a Neo4j graph application from scratch, this skill orchestrates provisioning, graph modeling, data loading, and query generation to produce a runnable artifact like a Jupyter notebook, Streamlit dashboard, or FastAPI backend.

GraphRAG document ingestion with Neo4j works by optionally processing document data during the loading stage to populate the graph, which is then queried using validated Cypher queries to support retrieval-augmented generation pipelines.

Yes, you can model a graph schema and apply constraints on Aura Free, as the skill supports schema-first modeling, constraint setup, and database provisioning across Aura, local Docker, and Desktop environments.

The best way to generate Cypher queries for a new Neo4j project is to use an orchestrated skill that validates CYPHER 25 syntax, enforces data presence success gates, and creates a dedicated query library.

You do not need Docker exclusively, as the skill supports connecting to existing databases or provisioning new ones on Aura Free or Desktop, allowing you to load synthetic or demo data without a local Docker setup.

Limitations when importing document data include enforcing success gates for schema completeness and data presence, requiring resumable stage execution, and needing a root SKILL.md YAML frontmatter to orchestrate the GraphRAG ingestion pipeline.