Knowledge Graph

Knowledge Graph

Build interconnected knowledge with entities and relationships. The knowledge graph enables complex queries, reasoning, and discovery of hidden connections.

Key Concepts

Entities

Named objects like people, companies, concepts, or topics extracted from your data.

Relationships

Typed connections between entities like "works_at", "mentions", "related_to".

Traversal

Query across relationships to discover multi-hop connections and patterns.

Automatic Entity Extraction

When you add memories or documents, Hebbrix automatically:

Extracts named entities (people, organizations, locations, concepts)
Creates relationships between entities based on context
Links entities to source memories for provenance tracking
Merges duplicate entities intelligently

Endpoints

Code Examples

Explore Entities

Python

from hebbrix import Hebbrix

client = Hebbrix()

# List all people in the knowledge graph
people = client.graph.entities.list(type="person", limit=20)

for person in people:
    print(f"{person.name} - {len(person.relationships)} connections")

Query Relationships

Python (Relationships)

# Find all entities connected to a person
entity = client.graph.entities.get("entity_abc123")

for rel in entity.relationships:
    print(f"{entity.name} --[{rel.type}]--> {rel.target.name}")

# Example output:
# John Doe --[works_at]--> Acme Inc
# John Doe --[knows]--> Jane Smith
# John Doe --[interested_in]--> Machine Learning

Graph Traversal

Python (Traversal)

# Find connections up to 2 hops away
results = client.graph.query(
    start_entity="John Doe",
    relationship_types=["works_at", "knows"],
    max_depth=2
)

# Discover hidden connections
for path in results.paths:
    print(" -> ".join([node.name for node in path]))

# Example output:
# John Doe -> Acme Inc -> Jane Smith
# John Doe -> Bob Wilson -> TechCorp

cURL Example

POST/v1/graph/query

curl -X POST "https://api.hebbrix.com/v1/graph/query" \
  -H "Authorization: Bearer YOUR_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
  "start_entity": "John Doe",
  "relationship_types": [
    "works_at",
    "knows"
  ],
  "max_depth": 2,
  "limit": 50
}'

Common Entity Types

person

organization

location

concept

product

event

technology

custom