neo4j-document-import-skill

Name: neo4j-document-import-skill
Author: neo4j-contrib

395installs33forks

How to install neo4j-document-import-skill?

neo4j-contrib/neo4j-skills/neo4j-document-import-skill

$npx -y skills add neo4j-contrib/neo4j-skills --skill neo4j-document-import-skill

Installs into the current project.

Summary

Ingests unstructured and semi-structured documents into Neo4j as a knowledge graph.

Files · 1

View on GitHub

SKILL.md

1	# Neo4j Document Import Skill
2
3	## When to Use
4
5	- Ingesting PDFs, HTML, plain text, Markdown into Neo4j as a knowledge graph
6	- Chunking documents and storing `:Chunk` nodes with embeddings
7	- Extracting entities and relationships from text with an LLM
8	- Using `SimpleKGPipeline` (neo4j-graphrag) programmatically
9	- Using Neo4j LLM Graph Builder (no-code web UI)
10	- Loading semi-structured JSON via `apoc.load.json`
11	- Connecting LangChain or LlamaIndex document loaders to Neo4j
12
13	## When NOT to Use
14
15	- Structured CSV / relational data → `neo4j-import-skill`
16	- GraphRAG retrieval after ingestion → `neo4j-graphrag-skill`
17	- Vector index creation → `neo4j-vector-search-skill`
18	- Cypher query writing → `neo4j-cypher-skill`
19
20	---
21
22	## Approach Decision Table
23
24	\| Situation \| Approach \|
25	\|---\|---\|
26	\| No code; drag-and-drop UX wanted \| LLM Graph Builder web UI \|
27	\| Programmatic pipeline; PDFs/text \| `SimpleKGPipeline` (neo4j-graphrag) \|
28	\| JSON / REST API responses \| `apoc.load.json` or Python + UNWIND \|
29	\| LangChain already in stack \| `Neo4jGraph` + document loader \|
30	\| LlamaIndex already in stack \| `Neo4jQueryEngine` / `Neo4jVectorStore` \|
31	\| Chunk-only (no entity extraction) \| Manual chunking + MERGE pattern \|
32
33	---
34
35	## Install
36
37	```bash
38	pip install neo4j-graphrag # includes SimpleKGPipeline
39	pip install neo4j-graphrag[openai] # + OpenAI LLM/embedder
40	pip install neo4j-graphrag[anthropic] # + Anthropic Claude
41	pip install neo4j-graphrag[google] # + Vertex AI / Gemini
42	pip install neo4j-graphrag[bedrock] # + Amazon Bedrock (boto3) — added v1.15.0
43	pip install neo4j-graphrag[ollama] # + Ollama (local)
44	pip install neo4j-graphrag[mistralai] # + MistralAI
45	pip install neo4j-graphrag[fuzzy-matching] # + FuzzyMatchResolver (rapidfuzz)
46	# spaCy entity resolver (Python <= 3.13 only — unsupported on 3.14+):
47	pip install neo4j-graphrag[nlp]
48	```
49
50	Requires: `neo4j>=5.17.0` (driver 6.x supported), Python>=3.10, Neo4j>=5.18.1 (Aura>=5.18.0).
51
52	---
53
54	## Step 1 — Define Graph Schema
55
56	Schema controls what the LLM extracts. Define before pipeline construction.
57
58	```python
59	# Option A — Simple string lists (LLM infers descriptions)
60	entities = ["Person", "Organization", "Location", "Product", "Event"]
61	relations = ["WORKS_AT", "LOCATED_IN", "KNOWS", "MENTIONS", "PART_OF"]
62	patterns = [
63	("Person", "WORKS_AT", "Organization"),
64	("Organization", "LOCATED_IN", "Location"),
65	("Person", "KNOWS", "Person"),
66	("Article", "MENTIONS", "Organization"),
67	]
68
69	# Option B — Rich GraphSchema (production; best extraction quality)
70	from neo4j_graphrag.experimental.components.schema import (
71	GraphSchema, NodeType, RelationshipType, PropertyType, ConstraintType
72	)
73	schema = GraphSchema(
74	node_types=[
75	NodeType(
76	label="Person",
77	description="A human individual",
78	properties=[
79	PropertyType(name="name", type="STRING"),
80	PropertyType(name="role", type="STRING"),
81	],
82	),
83	NodeType(
84	label="Organization",
85	description="A company or institution",
86	properties=[
87	PropertyType(name="name", type="STRING"),
88	PropertyType(name="industry", type="STRING"),
89	],
90	),
91	],
92	relationship_types=[
93	RelationshipType(label="WORKS_AT", description="Employment relationship"),
94	],
95	patterns=[("Person", "WORKS_AT", "Organization")],
96	# Optional: constraints emitted to ParquetWriter metadata (v1.15.0+)
97	constraints=[
98	ConstraintType(label="Person", property_name="name", type="UNIQUENESS"),
99	ConstraintType(label="Organization", property_name="name", type="KEY"),
100	],
101	)
102
103	# Option C — Auto-extract schema from text (no constraints)
104	schema = "EXTRACTED" # LLM infers types; noisier output
105	schema = "FREE" # No schema guidance; most noise
106	```
107
108	Use Option B for production; Option A for prototyping; `"EXTRACTED"` only for exploration.
109
110	---
111
112	## Step 2 — SimpleKGPipeline Setup
113
114	```python
115	import asyncio
116	from neo4j import GraphDatabase
117	from

People who liked this also viewed

6 picks

Type

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 2026471k1.3k
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 2026302k2.4k
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 2026179k2.4k
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 202690k39
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 202689k39
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 202689k39

neo4j-document-import-skill

395installs33forks

View on GitHub

How to install neo4j-document-import-skill?

neo4j-contrib/neo4j-skills/neo4j-document-import-skill

$npx -y skills add neo4j-contrib/neo4j-skills --skill neo4j-document-import-skill

Installs into the current project.

Summary

Ingests unstructured and semi-structured documents into Neo4j as a knowledge graph.

Files · 1

View on GitHub

SKILL.md

1	# Neo4j Document Import Skill
2
3	## When to Use
4
5	- Ingesting PDFs, HTML, plain text, Markdown into Neo4j as a knowledge graph
6	- Chunking documents and storing `:Chunk` nodes with embeddings
7	- Extracting entities and relationships from text with an LLM
8	- Using `SimpleKGPipeline` (neo4j-graphrag) programmatically
9	- Using Neo4j LLM Graph Builder (no-code web UI)
10	- Loading semi-structured JSON via `apoc.load.json`
11	- Connecting LangChain or LlamaIndex document loaders to Neo4j
12
13	## When NOT to Use
14
15	- Structured CSV / relational data → `neo4j-import-skill`
16	- GraphRAG retrieval after ingestion → `neo4j-graphrag-skill`
17	- Vector index creation → `neo4j-vector-search-skill`
18	- Cypher query writing → `neo4j-cypher-skill`
19
20	---
21
22	## Approach Decision Table
23
24	\| Situation \| Approach \|
25	\|---\|---\|
26	\| No code; drag-and-drop UX wanted \| LLM Graph Builder web UI \|
27	\| Programmatic pipeline; PDFs/text \| `SimpleKGPipeline` (neo4j-graphrag) \|
28	\| JSON / REST API responses \| `apoc.load.json` or Python + UNWIND \|
29	\| LangChain already in stack \| `Neo4jGraph` + document loader \|
30	\| LlamaIndex already in stack \| `Neo4jQueryEngine` / `Neo4jVectorStore` \|
31	\| Chunk-only (no entity extraction) \| Manual chunking + MERGE pattern \|
32
33	---
34
35	## Install
36
37	```bash
38	pip install neo4j-graphrag # includes SimpleKGPipeline
39	pip install neo4j-graphrag[openai] # + OpenAI LLM/embedder
40	pip install neo4j-graphrag[anthropic] # + Anthropic Claude
41	pip install neo4j-graphrag[google] # + Vertex AI / Gemini
42	pip install neo4j-graphrag[bedrock] # + Amazon Bedrock (boto3) — added v1.15.0
43	pip install neo4j-graphrag[ollama] # + Ollama (local)
44	pip install neo4j-graphrag[mistralai] # + MistralAI
45	pip install neo4j-graphrag[fuzzy-matching] # + FuzzyMatchResolver (rapidfuzz)
46	# spaCy entity resolver (Python <= 3.13 only — unsupported on 3.14+):
47	pip install neo4j-graphrag[nlp]
48	```
49
50	Requires: `neo4j>=5.17.0` (driver 6.x supported), Python>=3.10, Neo4j>=5.18.1 (Aura>=5.18.0).
51
52	---
53
54	## Step 1 — Define Graph Schema
55
56	Schema controls what the LLM extracts. Define before pipeline construction.
57
58	```python
59	# Option A — Simple string lists (LLM infers descriptions)
60	entities = ["Person", "Organization", "Location", "Product", "Event"]
61	relations = ["WORKS_AT", "LOCATED_IN", "KNOWS", "MENTIONS", "PART_OF"]
62	patterns = [
63	("Person", "WORKS_AT", "Organization"),
64	("Organization", "LOCATED_IN", "Location"),
65	("Person", "KNOWS", "Person"),
66	("Article", "MENTIONS", "Organization"),
67	]
68
69	# Option B — Rich GraphSchema (production; best extraction quality)
70	from neo4j_graphrag.experimental.components.schema import (
71	GraphSchema, NodeType, RelationshipType, PropertyType, ConstraintType
72	)
73	schema = GraphSchema(
74	node_types=[
75	NodeType(
76	label="Person",
77	description="A human individual",
78	properties=[
79	PropertyType(name="name", type="STRING"),
80	PropertyType(name="role", type="STRING"),
81	],
82	),
83	NodeType(
84	label="Organization",
85	description="A company or institution",
86	properties=[
87	PropertyType(name="name", type="STRING"),
88	PropertyType(name="industry", type="STRING"),
89	],
90	),
91	],
92	relationship_types=[
93	RelationshipType(label="WORKS_AT", description="Employment relationship"),
94	],
95	patterns=[("Person", "WORKS_AT", "Organization")],
96	# Optional: constraints emitted to ParquetWriter metadata (v1.15.0+)
97	constraints=[
98	ConstraintType(label="Person", property_name="name", type="UNIQUENESS"),
99	ConstraintType(label="Organization", property_name="name", type="KEY"),
100	],
101	)
102
103	# Option C — Auto-extract schema from text (no constraints)
104	schema = "EXTRACTED" # LLM infers types; noisier output
105	schema = "FREE" # No schema guidance; most noise
106	```
107
108	Use Option B for production; Option A for prototyping; `"EXTRACTED"` only for exploration.
109
110	---
111
112	## Step 2 — SimpleKGPipeline Setup
113
114	```python
115	import asyncio
116	from neo4j import GraphDatabase
117	from

People who liked this also viewed

6 picks

Type