neo4j-vector-index-skill
installsforks
How to install neo4j-vector-index-skill?
neo4j-contrib/neo4j-skills/neo4j-vector-index-skill
$npx -y skills add neo4j-contrib/neo4j-skills --skill neo4j-vector-index-skillCreate and manage Neo4j vector indexes, run vector similarity search (ANN/kNN),
SKILL.md
| 1 | ## When to Use |
| 2 | - Creating a vector index (`CREATE VECTOR INDEX`) on nodes or relationships |
| 3 | - Running vector similarity / nearest-neighbor search |
| 4 | - Storing embeddings on graph nodes during ingestion |
| 5 | - Indexing/querying embeddings already written by GDS algorithms |
| 6 | - Choosing similarity function, dimensions, HNSW params, or quantization |
| 7 | - Using `SEARCH` clause (2026.01+) or `db.index.vector.queryNodes()` (2025.x) |
| 8 | - Batch-updating embeddings after model change |
| 9 | - Combining vector results with immediate graph neighborhood (full retrieval_query pipelines → `neo4j-graphrag-skill`) |
| 10 | - Hybrid search that combines vector results with fulltext or other ranked sources |
| 11 | |
| 12 | ## When NOT to Use |
| 13 | - **GraphRAG pipelines** (VectorCypherRetriever, HybridCypherRetriever, retrieval_query) → `neo4j-graphrag-skill` |
| 14 | - **Fulltext-only / keyword-only search** (FULLTEXT INDEX, `db.index.fulltext.queryNodes`) → `neo4j-cypher-skill` |
| 15 | - **Computing GDS graph embeddings** (FastRP, Node2Vec, GraphSAGE) → `neo4j-gds-skill` |
| 16 | - **Index admin** (list all indexes, drop range/text/lookup indexes) → `neo4j-cypher-skill` |
| 17 | |
| 18 | --- |
| 19 | |
| 20 | ## Pre-flight — Determine Version |
| 21 | |
| 22 | Drives syntax choice: |
| 23 | ```cypher |
| 24 | CALL dbms.components() YIELD versions RETURN versions[0] AS neo4j_version |
| 25 | ``` |
| 26 | |
| 27 | | Version | Use | |
| 28 | |---|---| |
| 29 | | `2026.01` or higher | `SEARCH` clause (in-index filtering, preferred) | |
| 30 | | `2025.x` | `db.index.vector.queryNodes()` procedure (**deprecated 2026.04** — use `SEARCH` when on 2026.x) | |
| 31 | |
| 32 | --- |
| 33 | |
| 34 | ## Step 1 — Create Vector Index |
| 35 | |
| 36 | Node index (single label): |
| 37 | ```cypher |
| 38 | CYPHER 25 |
| 39 | CREATE VECTOR INDEX chunk_embedding IF NOT EXISTS |
| 40 | FOR (c:Chunk) ON (c.embedding) |
| 41 | OPTIONS { |
| 42 | indexConfig: { |
| 43 | `vector.dimensions`: 1536, |
| 44 | `vector.similarity_function`: 'cosine', |
| 45 | `vector.quantization.type`: 'SCALAR', |
| 46 | `vector.hnsw.m`: 16, |
| 47 | `vector.hnsw.ef_construction`: 100 |
| 48 | } |
| 49 | } |
| 50 | ``` |
| 51 | |
| 52 | Node index **with filterable properties** [2026.01+] — `WITH` declares which properties can be used in `SEARCH ... WHERE`: |
| 53 | ```cypher |
| 54 | CYPHER 25 |
| 55 | CREATE VECTOR INDEX chunk_embedding IF NOT EXISTS |
| 56 | FOR (c:Chunk) ON (c.embedding) |
| 57 | WITH [c.source, c.lang, c.published_year] // stored as metadata; filterable in SEARCH WHERE |
| 58 | OPTIONS { indexConfig: { `vector.dimensions`: 1536, `vector.similarity_function`: 'cosine' } } |
| 59 | ``` |
| 60 | |
| 61 | Multi-label index with filterable properties [2026.01+]: |
| 62 | ```cypher |
| 63 | CYPHER 25 |
| 64 | CREATE VECTOR INDEX doc_embedding IF NOT EXISTS |
| 65 | FOR (n:Document|Article) ON n.embedding |
| 66 | WITH [n.author, n.published_year, n.lang] |
| 67 | OPTIONS { indexConfig: { `vector.dimensions`: 1536, `vector.similarity_function`: 'cosine' } } |
| 68 | ``` |
| 69 | |
| 70 | Relationship index: |
| 71 | ```cypher |
| 72 | CYPHER 25 |
| 73 | CREATE VECTOR INDEX rel_embedding IF NOT EXISTS |
| 74 | FOR ()-[r:HAS_CHUNK]-() ON (r.embedding) |
| 75 | OPTIONS { indexConfig: { `vector.dimensions`: 768, `vector.similarity_function`: 'cosine' } } |
| 76 | ``` |
| 77 | |
| 78 | **`WITH` property types** — only scalar types allowed: `INTEGER`, `FLOAT`, `STRING`, `BOOLEAN`, `DATE`, `ZONED DATETIME`, `LOCAL DATETIME`, `ZONED TIME`, `LOCAL TIME`, `DURATION`. Not allowed: `LIST`, `POINT`, or the vector property itself. |
| 79 | |
| 80 | **Index config reference:** |
| 81 | |
| 82 | | Parameter | Type | Default | Notes | |
| 83 | |---|---|---|---| |
| 84 | | `vector.dimensions` | INTEGER 1–4096 | none | Required; must match embedding model exactly | |
| 85 | | `vector.similarity_function` | STRING | `'cosine'` | `'cosine'` or `'euclidean'` | |
| 86 | | `vector.quantization.type` | STRING | `'SCALAR'` | `NONE`, `SCALAR`, or `BINARY` [2026.06+]; reduces storage; BINARY smallest, most aggressive; needs vector-2.0+ (5.18+) | |
| 87 | | `vector.quantization.enabled` | BOOLEAN | `true` | **Deprecated 2026.06** — use `vector.quantization.type` | |
| 88 | | `vector.default_search_expansion_factor` | FLOAT | `1.0` NONE / `1.5` SCALAR / `2.0` BINARY | [2026.06+]; value >1.0 on quantized vectors enables automatic rescoring with full-precision vectors; not settable at query time | |
| 89 | | `vector.hnsw.m` | INTEGER 1–512 | `16` | HNSW graph connections; higher = better recall, more memory | |
| 90 | | `vector.hnsw.ef_co |
People who liked this also viewed
6 picksType
- azure-storageAzure Storage Services including Blob Storage, File Shares, Queue Storage, Table Storage, and Data Lake. Answers questions about storage access tiers (hot, cool, cold, archive), when to use each tier,SkillsJul 2026
- supabase-postgres-best-practicesPostgres performance optimization and best practices from Supabase. Use this skill when writing, reviewing, or optimizing Postgres queries, schema designs, or database configurations.SkillsJul 2026
- supabaseUse when doing ANY task involving Supabase. Triggers: Supabase products (Database, Auth, Edge Functions, Realtime, Storage, Vectors, Cron, Queues); client libraries and SSR integrations (supabase-js, SkillsJul 2026
- convex-quickstartCreates or adds Convex to an app. Use for new Convex projects, npm create convex@latest, frontend setup, env vars, or the first npx convex dev run.SkillsJul 2026
- convex-performance-auditAudits Convex performance for reads, subscriptions, write contention, and function limits. Use for slow features, insights findings, OCC conflicts, or read amplification.SkillsJul 2026
- convex-setup-authSets up Convex auth, identity mapping, and access control. Use for login, auth providers, users tables, protected functions, or roles in a Convex app.SkillsJul 2026
