{
  "uid": "cogitave.learn.core-model-fundamentals.introduction",
  "kind": "moduleUnit",
  "href": "/modules/core-model-fundamentals/introduction/",
  "title": "Introduction",
  "summary": "",
  "type": null,
  "products": [],
  "roles": [],
  "levels": [],
  "subjects": [],
  "headings": [
    "What this module teaches",
    "Status: read this before you build against it"
  ],
  "source": "Ask five different systems where a piece of company knowledge lives - the docs\nsite, the internal developer portal, the compliance evidence store, the infra\ninventory, an agent's retrieval index - and a normal company gives you five\ndifferent answers, in five different stores, that drift from each other the\nmoment two of them are updated separately.\n\nAn AI-native organization refuses that split. It runs on a **single typed\nproperty graph** that all of those surfaces are *projections* of - not synced\ncopies, not separate stores that happen to agree today. Docs, the internal\ndeveloper portal, governance and compliance evidence, infrastructure inventory,\nand agent knowledge all read from the same graph, through the same query layer,\nand that layer is MCP-native: a human using the API/UI and an agent using MCP\nget *the same knowledge from the same model*. **Cogitave's Core** is one such\ngraph - the worked example this module builds on. Its canonical\n[architecture doc](../../../../core/docs/architecture.md) states the same thesis\nfor that instance, and this module - and the rest of this path - builds directly\non it.\n\n## What this module teaches\n\nTwo ideas make the one-graph claim more than a slogan:\n\n- **The one-model thesis** - why one graph, not one store per domain, and what\n  that buys in exchange for the discipline it demands.\n- **The property-graph substrate** - why that one graph is specifically a\n  *labeled property graph*, what a design decision like\n  [Cogitave's ADR-0001](../../../../core/docs/decisions/0001-property-graph-as-substrate.md)\n  accepted and rejected to get there, and what makes it fast enough to serve.\n\n## Status: read this before you build against it\n\nCogitave's architecture doc opens by calling itself a \"Day 0 canonical spec.\"\nTreat it that way here, too. A one-graph model and the decision behind it are\nthings you **decide** - accept, not merely propose - and this module teaches\nCogitave's as documented decisions. It is not a claim that every product has\nfinished projecting into the graph, and it is not a hands-on module: you will\nnot stand up a canonical model here, you will learn to read the spec that\ngoverns one, the same spec an auditor or a querying agent relies on.\n",
  "partOf": "cogitave.learn.core-model-fundamentals",
  "durationInMinutes": 4,
  "quiz": null
}