{
  "id": "knowledge-fabric-impl:mempalace-fabric",
  "_kind": "KnowledgeFabricImpl",
  "_file": "agent-stack/knowledge-fabric-impls/community-knowledge-fabrics.yaml",
  "_cluster": "agent-stack",
  "attributes": {
    "displayName": "MemPalace as Agent Memory Fabric",
    "description": "MemPalace as a benchmarked agent memory knowledge fabric. Provides\npersistent memory with quality scoring and retrieval ranking. Memories\nare stored with relevance scores that decay or strengthen over time\nbased on usage patterns. The benchmarking focus means MemPalace\nprioritizes measurable retrieval accuracy over feature breadth. As a\nknowledge fabric, MemPalace bridges ephemeral conversation context\nand persistent organizational knowledge through scored, retrievable\nmemory entries.\n",
    "knowledgeFileFormats": [
      "structured-memories",
      "scored-entries"
    ],
    "retrievalStrategy": "scored-retrieval",
    "knowledgePersistence": "memory-service",
    "knowledgeScopes": [
      "agent",
      "user"
    ],
    "autoExtractionSupport": true,
    "notes": "MemPalace's key differentiator is its benchmarking focus — it is\nevaluated against other memory systems for accuracy, recall, and\nefficiency. This makes it a strong choice for agents where memory\nreliability is critical and needs to be measured, rather than\nassumed.\n"
  },
  "outgoingEdges": [
    {
      "from": "knowledge-fabric-impl:mempalace-fabric",
      "to": "layer:12-knowledge-fabric",
      "kind": "realizes",
      "attributes": {}
    },
    {
      "from": "knowledge-fabric-impl:mempalace-fabric",
      "to": "tool:mempalace",
      "kind": "integrates_with",
      "attributes": {}
    },
    {
      "from": "knowledge-fabric-impl:mempalace-fabric",
      "to": "memory-system:mempalace-service",
      "kind": "uses_memory_system",
      "attributes": {}
    }
  ],
  "incomingEdges": []
}