neo4j-snowflake-graph-analytics-skill
Run Neo4j Graph Analytics algorithms (PageRank, Louvain, WCC, Dijkstra, KNN, Node2Vec, FastRP, GraphSAGE) directly inside Snowflake without moving data. Use when running graph algorithms against Snowflake tables via the Neo4j Snowflake Native App ("GDS Snowflake", "graph algorithms in Snowflake", "Neo4j Graph Analytics"). Covers installation, privilege setup, project-compute-write pattern, and SQL CALL syntax. Does NOT cover Cypher or Neo4j DBMS queries — use neo4j-cypher-skill. Does NOT cover Aura Graph Analytics — use neo4j-aura-graph-analytics-skill. Does NOT cover self-managed GDS — use neo4j-gds-skill.
88
82%
Does it follow best practices?
Impact
100%
1.36xAverage score across 3 eval scenarios
Passed
No known issues
Quality
Discovery
100%Based on the skill's description, can an agent find and select it at the right time? Clear, specific descriptions lead to better discovery.
This is an excellent skill description that hits all dimensions strongly. It provides specific algorithms and actions, includes rich natural trigger terms, explicitly answers both what and when, and proactively distinguishes itself from related skills with clear negative boundaries. The description is concise yet comprehensive, serving as a near-ideal example of skill description writing.
| Dimension | Reasoning | Score |
|---|---|---|
Specificity | Lists multiple specific algorithms (PageRank, Louvain, WCC, Dijkstra, KNN, Node2Vec, FastRP, GraphSAGE) and concrete actions (installation, privilege setup, project-compute-write pattern, SQL CALL syntax). Very concrete and actionable. | 3 / 3 |
Completeness | Clearly answers 'what' (run graph analytics algorithms in Snowflake, covers installation/setup/patterns) and 'when' (explicit 'Use when' clause with trigger terms). Also includes explicit negative boundaries distinguishing from related skills, which strengthens the 'when' guidance. | 3 / 3 |
Trigger Term Quality | Includes excellent natural trigger terms: 'Neo4j', 'Snowflake', 'graph algorithms', 'PageRank', 'Louvain', 'GDS Snowflake', 'Neo4j Graph Analytics', 'Neo4j Snowflake Native App', and specific algorithm names users would naturally mention. | 3 / 3 |
Distinctiveness Conflict Risk | Exceptionally distinctive — explicitly delineates boundaries with three related skills (neo4j-cypher-skill, neo4j-aura-graph-analytics-skill, neo4j-gds-skill) using 'Does NOT cover' clauses, making conflict risk very low. | 3 / 3 |
Total | 12 / 12 Passed |
Implementation
64%Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.
This is a solid, actionable skill with executable SQL examples and clear conceptual framing (Project → Compute → Write). Its main weaknesses are the inconsistent placeholder names between the privilege setup code and subsequent examples (MY_CONSUMER_ROLE vs GRAPH_USER_ROLE), the lack of validation checkpoints in workflows, and the large inline algorithm reference tables that could benefit from being split into a separate file for better progressive disclosure.
Suggestions
Fix the placeholder name inconsistency: the privilege setup uses MY_CONSUMER_ROLE/MY_DATABASE/MY_SCHEMA but the full example uses GRAPH_USER_ROLE/P2P/PUBLIC — unify these or add a clear mapping note.
Add explicit validation steps to the workflow, e.g., checking that the output table was created and contains expected results before proceeding to chaining.
Move the exhaustive algorithm tables to a separate ALGORITHMS.md reference file and keep only the most common 5-6 algorithms inline with a link to the full list.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | Generally efficient but includes some content that could be trimmed — the exhaustive algorithm tables (30+ algorithms) add significant length and could be referenced externally. The privilege setup SQL block also has a mismatch between placeholder names in the code vs. the note below it (MY_CONSUMER_ROLE vs GRAPH_USER_ROLE). However, most sections are well-structured and avoid explaining concepts Claude already knows. | 2 / 3 |
Actionability | Provides fully executable SQL code for privilege setup, algorithm invocation, view creation, and result inspection. The full WCC example is copy-paste ready with clear parameter explanations, and the configuration reference sections give concrete JSON structures. | 3 / 3 |
Workflow Clarity | The Project → Compute → Write pattern is clearly articulated, and the full example demonstrates the workflow well. However, there are no explicit validation checkpoints — no step to verify the projection succeeded, no error handling guidance within the workflow, and the chaining example lacks verification between steps. The placeholder name mismatch (MY_CONSUMER_ROLE in code vs GRAPH_USER_ROLE in the note and later example) could cause real errors. | 2 / 3 |
Progressive Disclosure | The skill has good external links to documentation for further reading, but the algorithm tables (which take up a large portion of the file) could be split into a separate reference file. With no bundle files, all content is inline in a single document that's quite long. The structure within the file is well-organized with clear headers, but the monolithic nature hurts discoverability. | 2 / 3 |
Total | 9 / 12 Passed |
Validation
90%Checks the skill against the spec for correct structure and formatting. All validation checks must pass before discovery and implementation can be scored.
Validation — 10 / 11 Passed
Validation for skill structure
| Criteria | Description | Result |
|---|---|---|
frontmatter_unknown_keys | Unknown frontmatter key(s) found; consider removing or moving to metadata | Warning |
Total | 10 / 11 Passed | |
- Repository
- neo4j-contrib/neo4j-skills
- Commit
66ed0e1
- Reviewed
Table of Contents
Is this your skill?
If you maintain this skill, you can claim it as your own. Once claimed, you can manage eval scenarios, bundle related skills, attach documentation or rules, and ensure cross-agent compatibility.