Skip to main content
Autonomous agents write new facts into the graph. Agents that process data enable the world to evolve. Worlds supports multiple strategies for ingesting new knowledge and continuously updating data within an isolated world.

​
SPARQL updates

A world mutates its graph state using RDF patches (granular additions and deletions of facts). Because Worlds functions internally as a chronological, append-only ledger, these discrete patches permanently preserve historical truth and temporal evolution. However, agents and developers rarely write these patches manually. The most common way to mutate graph state is executing SPARQL updates. When an agent runs an INSERT or DELETE command, Worlds translates it into a deterministic RDF patch and applies it. 
PREFIX schema: <http://schema.org/>

INSERT DATA {
  <user:ethan> schema:worksAt <org:wazoo> .
}

​
Ingestion strategies

​
One-time import

Use the CLI or SDK to perform a bulk import of existing RDF data. This is ideal for bootstrapping a new world with a base ontology or research dataset. 
# Import a Turtle file into a world
worlds import --world <world-id> --file ./ontology.ttl

​
Direct SDK ingestion

For real-time applications, use @wazoo/worlds-sdk to insert data as it is generated by your agentic loops.

​
Data synchronization

To keep your world memory synced with external operations, use a forward-sync proxy to hook into webhooks exposed by your services e.g., repository events or team communication streams.
  1. Map external events: Capture the real-time webhook payloads broadcast by your source system.
  2. Translate to facts: Convert the resulting JSON data into structured graph mutations. For example, translate a new repository ticket event into a SPARQL INSERT operation asserting that a specific user opened it.
  3. Apply updates: Execute the mutation against the world, allowing the engine to translate it into a safe, atomic RDF patch.