Introduction

In the first three modules of this path you learned Core's node and edge model, how to query it, and how to reach it over MCP. This module closes the loop on a different question: where does the content come from? A product spec, an engineering standard, an ADR, or the module you are reading right now - how does it end up as a node you can query, rather than a file that quietly drifts from whatever else says the same thing?

One graph, not a pile of silos

Cogitave Core's architecture states the thesis directly: docs, the internal developer portal, governance, compliance evidence, and infra are projections of one graph - never separate stores. A Doc, a Standard, an ADR, and this very learning Unit are all typed nodes in the same property graph, addressed the same way, queried through the same MCP tools a human's UI calls. There is no second copy for "the docs system" and a third for "the learn platform."

That only works if content actually gets into the graph, and a fact true in one place cannot silently go stale in another. This module covers both halves: how a document becomes a node (unit 2), and how Cogitave keeps one authoritative owner per load-bearing fact so the rest of the estate cites it instead of re-typing it and drifting (unit 3).

What you will get from this module

The ability to explain, concretely, how a document becomes a queryable node, and why "the estate is a second brain" is more than a slogan: it names a real gap - facts restated in prose instead of linked - and a real, checkable fix. This is the last module of Build on Cogitave Core; finishing it earns the path's trophy.