neo4j-gds-skill

What problem does it solve?

This Skill helps you run Neo4j Graph Data Science (GDS) graph algorithms reliably without manually wiring projections, choosing execution modes, or handling common GDS configuration pitfalls.

Core Features & Use Cases

• Graph projection and graph catalog management: Create native and Cypher-based projections, inspect projected graphs, estimate memory, list catalog graphs, and drop them when finished.
• Algorithm execution in the right mode: Use stream, stats, mutate, and write to control side effects and support pipelines like FastRP → KNN recommendations.
• Core graph analytics and embeddings: Run PageRank, Louvain/Leiden-style community workflows, WCC, Betweenness Centrality, Node Similarity, FastRP embeddings, and KNN similarity search, including writing embeddings for later vector indexing.
• Practical guardrails: Pre-flight checks (embedded plugin availability), compatibility expectations, and mitigations for common errors such as OOM, missing properties, or leftover catalog graphs.

Quick Start

Use the neo4j-gds-skill to project a graph and run PageRank on it, then write results back to Neo4j so they are queryable later.