Module 6 of 6 · 15 min

# Collate Agentic Data Platform

How context, the knowledge graph and the ontology explorer ground agents in trusted metadata.

## Why this matters

Ingestion, classification, lineage, quality and ownership together turn a set of tables into context an agent can reason over. This module shows how the knowledge graph and the ontology explorer use that context to ground agent answers.

## Before you start

### Curated metadata is the context

Every curation step covered so far contributes: a description gives an asset meaning, lineage gives it provenance, a quality result gives it a trust signal, an owner gives it accountability. Together they are the context an agent needs to answer correctly.

### A knowledge graph expresses relationships

An asset list names what exists. A graph describes how assets relate to each other and to business concepts. That structure is what allows a question posed in business language to reach the right physical asset.

### Ungoverned metadata produces confident wrong answers

The failure is an agent that answers fluently from stale, untested or mislabeled metadata.

## The interactive demo

Work through the demo below. It runs in your browser with nothing to install. Watch for:

*   •Using the AI native interface
*   •The Context Center that feeds the Knowledge Graph
*   •The Ontology Explorer
*   •A walkthrough of AI Studio
*   •The new Dashboards

This demo is also available on its own at [the Learning Center](/learning-center/product-demo-agentic-data-platform).

## After this module you should be able to

*   ✓Explain how curated metadata becomes usable context for an agent
*   ✓Describe what a knowledge graph adds beyond a list of assets
*   ✓Argue why governance quality determines AI answer quality