Memory

Provides a persistent memory knowledge graph with entities, relations, and observations that you can store, retrieve, and query across sessions.

python

0

GitHub Stars

python

Language

7 months ago

First Indexed

3 months ago

Catalog Refreshed

Documentation & install

Readme and setup notes from the catalogue, plus a client-ready config you can copy for your MCP host.

Installation

Add the following to your MCP client configuration file.

Configuration

View docs

{
  "mcpServers": {
    "jcdiv47-mcp-server-memory": {
      "command": "/path/to/uv",
      "args": [
        "--directory",
        "/path/to/mcp_server_memory/src",
        "run",
        "python",
        "-m",
        "mcp_server_memory"
      ],
      "env": {
        "MEMORY_FILE_PATH": "/path/to/memory.json"
      }
    }
  }
}

Memory MCP Server provides a Python-based MCP server that maintains a persistent knowledge graph to store, retrieve, and query entities, relations, and observations. This enables AI systems to remember information across sessions and perform structured memory operations efficiently.

How to use

You interact with the Memory MCP Server through an MCP client to manage memory entities, their relations, and observations. Use the available tools to create memory nodes, connect them with typed relations, add new observations, and query or retrieve nodes by name. You can also read the entire graph to inspect its current state and perform searches to locate nodes that match your queries.

Practical usage patterns include creating entities to represent concepts or memories, forming relations to define how those concepts relate, appending observations to capture new details over time, and deleting nodes or observations that are no longer relevant. To retrieve information, use search and read operations to locate and inspect specific nodes or segments of your knowledge graph.

Key operations you can perform with the MCP client include creating entities, creating relations between entities, adding observations to existing entities, deleting entities or observations, deleting relations, reading the full graph, searching for nodes, and opening specific nodes by name.

How to install

Prerequisites you need before starting are Python and a compatible MCP runner like your preferred MCP client or the UV-based runner. You should also be prepared to configure a memory file path where the knowledge graph will be stored.

Install the Memory MCP Server via UV (recommended):

uv add mcp-server-memory

Install the Memory MCP Server via pip:

pip install mcp-server-memory

Running the server

Run the server directly or with a custom memory file path. The server exposes a straightforward command you invoke from your shell.

Run the server directly

mcp-server-memory

Set a custom memory file path when starting the server

MEMORY_FILE_PATH=/path/to/memory.json mcp-server-memory

Available tools

create_entities

Create multiple new entities in the knowledge graph.

create_relations

Create multiple new relations between entities.

add_observations

Add new observations to existing entities.

delete_entities

Delete entities and their associated relations.

delete_observations

Delete specific observations from entities.

delete_relations

Delete specific relations from the graph.

read_graph

Read the entire knowledge graph.

search_nodes

Search for nodes matching a query.

open_nodes

Retrieve specific nodes by name.