{
  "id": "knowledge-fabric-impl:weaviate-fabric",
  "_kind": "KnowledgeFabricImpl",
  "_file": "agent-stack/knowledge-fabric-impls/vector-store-fabrics.yaml",
  "_cluster": "agent-stack",
  "attributes": {
    "displayName": "Weaviate as Knowledge Fabric",
    "description": "Weaviate as a schema-aware knowledge fabric. Combines vector, keyword, and\nhybrid BM25+vector search with built-in vectorization modules for text,\nimages, and multi-modal data. GraphQL and REST APIs provide flexible\nquerying. Weaviate's generative search modules pipe retrieved context\ndirectly into LLM prompts, making it a native RAG backend. Multi-tenancy,\nfiltering, and horizontal scaling support enterprise deployments. As a\nknowledge fabric, Weaviate uniquely combines structured schema awareness\nwith semantic search.\n",
    "knowledgeFileFormats": [
      "embeddings",
      "document-chunks",
      "structured-objects"
    ],
    "retrievalStrategy": "hybrid",
    "knowledgePersistence": "vector-store",
    "knowledgeScopes": [
      "organization",
      "enterprise"
    ],
    "autoExtractionSupport": false,
    "notes": "Weaviate's distinguishing feature as a knowledge fabric is its schema-aware\nobject model — knowledge is not just embedded vectors but typed objects with\nproperties and cross-references. This enables both semantic similarity search\nand structured filtering. The built-in generative search module (generate)\nmakes it a complete RAG backend without external orchestration.\n"
  },
  "outgoingEdges": [
    {
      "from": "knowledge-fabric-impl:weaviate-fabric",
      "to": "layer:12-knowledge-fabric",
      "kind": "realizes",
      "attributes": {}
    },
    {
      "from": "knowledge-fabric-impl:weaviate-fabric",
      "to": "tool:weaviate",
      "kind": "integrates_with",
      "attributes": {}
    }
  ],
  "incomingEdges": []
}