MCP Memory Server

Provides an on-device persistent memory store for MCP-enabled editors, with multi-project isolation and Markdown ingestion.

python

0

GitHub Stars

python

Language

4 months ago

First Indexed

2 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": {
    "iamjpsharma-mcpserver": {
      "command": "/ABSOLUTE/PATH/TO/mcp-memory-server/.venv/bin/python",
      "args": [
        "-m",
        "mcp_memory.server"
      ],
      "env": {
        "MCP_MEMORY_PATH": "/ABSOLUTE/PATH/TO/mcp-memory-server/mcp_memory_data"
      }
    }
  }
}

You can run a local, private memory server that stores contextual data for MCP-compatible editors like Windsurf and VS Code. It keeps your information on your machine, supports multiple projects, and ingests Markdown documents to enhance your editor’s memory capabilities.

How to use

To use the MCP Memory Server, you run it as a local process and configure your editor to connect via the MCP stdio transport. Once configured, you can ingest project files, then perform semantic memory searches and manually add memory fragments from within your editor. This lets the editor recall relevant context across your coding sessions.

How to install

Prerequisites you need before installation:

Python 3.8+ (with venv support)
pip with virtual environment tools (venv)

Step 1: Clone the project and set up the local memory server then install dependencies.

git clone https://github.com/iamjpsharma/MCPServer.git
cd MCPServer/mcp-memory-server

# Create and activate virtual environment
python3 -m venv .venv
source .venv/bin/activate

# Install dependencies
pip install -e .

Configure your editor to run the server

You configure your editor’s MCP client to start the memory server using the Python interpreter from your virtual environment. Use the absolute path to the project in your configuration.

{
  "mcpServers": {
    "memory": {
      "command": "/ABSOLUTE/PATH/TO/mcp-memory-server/.venv/bin/python",
      "args": ["-m", "mcp_memory.server"],
      "env": {
        "MCP_MEMORY_PATH": "/ABSOLUTE/PATH/TO/mcp-memory-server/mcp_memory_data"
      }
    }
  }
}

Usage actions once connected

After the server is running and your editor is connected, you can ingest files into memory, search for project content semantically, and add memory fragments manually to tailor the assistant’s long-term memory to your projects.

Health check

# From within the virtual environment
python -m mcp_memory.server --help

Available tools

memory_search

Performs semantic searches over ingested memory fragments for a given project ID and query.

memory_add

Manually add memory fragments to a specific project ID to enrich the editor's long-term memory.