Question 1

What is a context graph?

Accepted Answer

A context graph is a relational map of your business entities, their attributes, and the semantic links between them. Galaxy builds it from your modeled objects so AI can traverse real relationships — which customer owns which account, which metric depends on which source — instead of guessing. It replaces data silos with a shared context graph that teams and AI agents can query directly.

Question 2

What's the difference between a context graph and a knowledge graph?

Accepted Answer

They're closely related. A knowledge graph connects entities like customers, products, contracts, and systems to make context navigable. Galaxy's context graph is a governed knowledge graph purpose-built for AI runtime: it adds shared semantics, entity resolution, provenance, and permissions so agents retrieve not just similar text, but the right connected facts.

Question 3

How does the context graph improve AI answers?

Accepted Answer

By giving models an explicit, navigable structure of how business concepts relate, the context graph lets AI follow real relationships and provenance when answering, which sharply reduces hallucination compared with prompting over raw text or ungoverned vector retrieval. It combines RAG patterns with graph-based grounding so systems surface relevant documents and the relationships between them.

Question 4

Can I trace where an answer came from?

Accepted Answer

Yes. Every node and link in the Galaxy context graph carries lineage and provenance, so any AI answer can be traced back to the underlying sources, definitions, and records that produced it.

How the graph works in Galaxy