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

neo4j-gds-skill

Runs Neo4j Graph Data Science algorithms via Python client or Cypher on embedded GDS plugin deployments.

Updated Jul 24, 2026
Search Tags:#knn#neo4j#cypher#pagerank#graph data science#graph projection#fastrp
One-click install
npx skills add https://github.com/eklyukin/my-ai-config --skill neo4j-gds-skill-eklyukin
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: neo4j-gds-skill
Source: https://github.com/eklyukin/my-ai-config/tree/main/skills/neo4j-gds-skill
Command: npx skills add https://github.com/eklyukin/my-ai-config --skill neo4j-gds-skill-eklyukin

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill requires graphdatascience, and includes references (resource) components.

What problem does it solve? Running graph algorithms on Neo4j requires correct graph projection, memory estimation, execution mode selection, and version-aware API usage, and mistakes cause OOM errors, failed projections, or unqueryable results. This Skill provides verified patterns for the embedded GDS plugin so algorithms run correctly the first time. ## Core Features & Use Cases - Graph Projection & Catalog Management: Native and Cypher projection, undirected orientation, subgraph filtering, memory estimation, and graph drop/list operations. - Core Algorithms with Correct Modes: PageRank, Louvain, Leiden, WCC, Betweenness Centrality, Node Similarity, FastRP, and KNN with stream/stats/mutate/write guidance and v2 Python client endpoints. - Recommendation Pipelines: The FastRP → KNN pattern for building similarity relationships from node embeddings, ready for Neo4j vector indexes. - Use Case: A data engineer needs product recommendations from a co-purchase graph. The Skill projects the graph undirected, estimates memory, mutates FastRP embeddings, writes KNN similarity relationships, and drops the projection cleanly. ## Quick Start Ask the agent to project a Person KNOWS graph in Neo4j and run PageRank using the GDS Python client, writing the scores back to the database.

Frequently Asked Questions about neo4j-gds-skill

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

FAQPage Schema
How do I run PageRank on Neo4j with the GDS Python client?▼

Connect with GraphDataScience, project a named graph via gds.v2.graph.project, then call gds.v2.page_rank.stream to inspect scores or gds.v2.page_rank.write with write_property to persist them. Verify the plugin first with gds.server_version().

How to build a recommendation pipeline with FastRP and KNN in Neo4j?▼

Project the graph undirected, mutate FastRP embeddings with a fixed random_seed, then run KNN write with node_properties set to the embedding to create SIMILAR relationships. Drop the projection afterward to free JVM heap.

What is the difference between stream, mutate, and write modes in GDS?▼

Stream returns per-node rows without side effects, mutate adds properties to the in-memory graph only for chaining algorithms, and write persists results to the Neo4j database. The standard pattern is stream to verify, mutate to chain, write to persist.

Does the graphdatascience Python client work with Aura Graph Analytics sessions?▼

No, this Skill covers only the embedded GDS plugin on Aura Pro, self-managed, or local deployments. Aura Graph Analytics sessions, GdsSessions, and remote projection require the neo4j-aura-graph-analytics-skill instead.

Why does my GDS projection fail with insufficient heap memory?▼

The projected graph exceeds available JVM heap. Run gds.graph.project.estimate before projecting, compare requiredMemory against dbms.memory.heap.max_size, and either reduce the graph scope or increase heap allocation.

When should I use native projection versus Cypher projection in GDS?▼

Use native projection for simple label and relationship type selections, which is the shortest Python-client path. Use Cypher projection when you need filtering, transformations, computed properties, or heterogeneous projections that native projection cannot express.

Related Skills

Explore More Data & Analytics Skills→
neo4j-contribneo4j-contrib
101

neo4j-aura-graph-analytics-skill

Execute Neo4j Aura Graph Analytics workflows for graph projection, algorithm runs, and result retrieval.

Official
Advanced
FalkorDBFalkorDB
22

Create and query vector indexes

Create and query HNSW vector indexes in FalkorDB graph databases.

Official
Intermediate
MassimilianoPiliMassimilianoPili

dba-graphdb

Manage Neo4j 5+ and Apache AGE graph database design and administration tasks.

Community
Advanced
motherduckdbmotherduckdb
53

motherduck-build-dashboard

Compose multi-section MotherDuck analytics dashboards as saved Dives from live data.

Official
Advanced