io.github.Abhigyan-Shekhar/Waggle-mcp

io.github.Abhigyan-Shekhar/Waggle-mcp·v0.1.8·Memory & RAG

Quality Score

/100

Persistent graph-backed conversational memory for AI agents.

Repository →

§01 Install

Claude Desktop (uvx)

{
  "mcpServers": {
    "Waggle-mcp": {
      "command": "uvx",
      "args": [
        "waggle-mcp"
      ],
      "env": {
        "WAGGLE_TRANSPORT": "stdio",
        "WAGGLE_BACKEND": "sqlite",
        "WAGGLE_DB_PATH": "~/.waggle/memory.db",
        "WAGGLE_DEFAULT_TENANT_ID": "local-default",
        "WAGGLE_MODEL": "all-MiniLM-L6-v2"
      }
    }
  }
}

§02 Environment variables

WAGGLE_TRANSPORT

Transport mode for the MCP server.

WAGGLE_BACKEND

Backend database type: sqlite for local use or neo4j for service deployments.

WAGGLE_DB_PATH

Path to the SQLite memory database when WAGGLE_BACKEND is sqlite.

WAGGLE_DEFAULT_TENANT_ID

Default tenant ID for local or shared memory isolation.

WAGGLE_MODEL

Sentence-transformers model used for local embeddings.

§03 MCP Quality Score · methodology