mnema MCP配置与使用教程

{
  "id": "ai.zimac/mnema",
  "type": "mcp-server",
  "version": "0.2.0",
  "displayName": "mnema",
  "description": "Local-first, formally verified memory for AI agents: consult/learn/related/list over MCP.",
  "repository": {
    "url": "https://github.com/Zimac-AI/mnema",
    "source": "github"
  },
  "homepage": "https://mnema.zimac.ai",
  "distribution": {
    "packages": [
      {
        "registryType": "mcpb",
        "identifier": "https://github.com/Zimac-AI/mnema/releases/download/v0.2.0/mnema-0.2.0-macos-universal.mcpb",
        "version": "0.2.0",
        "transport": "stdio",
        "environmentVariables": [
          {
            "name": "MNEMA_DATA_DIR",
            "description": "Where the store lives (default ~/.mnema)."
          },
          {
            "name": "MNEMA_URL",
            "description": "Target store URL. Loopback (default http://127.0.0.1:7077) auto-starts a local daemon; a remote URL makes the server a pure authenticated bridge."
          },
          {
            "name": "MNEMA_TOKEN",
            "description": "Bearer credential when MNEMA_URL points at a token-protected remote store.",
            "isSecret": true
          }
        ]
      },
      {
        "registryType": "mcpb",
        "identifier": "https://github.com/Zimac-AI/mnema/releases/download/v0.2.0/mnema-0.2.0-linux-x86_64.mcpb",
        "version": "0.2.0",
        "transport": "stdio"
      }
    ],
    "remotes": []
  },
  "dependencies": [],
  "installTargets": [
    "claude-code",
    "claude-desktop",
    "cursor",
    "vscode",
    "trae",
    "cherry-studio",
    "lingma",
    "windsurf",
    "cline",
    "workbuddy"
  ],
  "keywords": [],
  "provenance": {
    "origin": "official-mcp-registry",
    "originalId": "ai.zimac/mnema",
    "originalUrl": "https://registry.modelcontextprotocol.io/v0.1/servers/ai.zimac%2Fmnema/versions/latest",
    "isOfficial": true,
    "status": "active"
  }
}