Work Memory

Persistent, session-aware memory management MCP server for cross-tool context and fast recall.

typescript

14

GitHub Stars

typescript

Language

5 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": {
    "moontmsai-work-memory-mcp": {
      "command": "node",
      "args": [
        "/PATH/work-memory/dist/index.js"
      ],
      "env": {
        "NODE_ENV": "production",
        "LOG_LEVEL": "WARN",
        "DB_FILENAME": "database.sqlite",
        "WORK_MEMORY_DIR": "/PATH/work-memory/data/"
      }
    }
  }
}

You run a dedicated MCP server that manages work memories and shares context across AI tools. It stores memories, todos, and session data in a persistent database, enabling seamless continuity as you move between tools like Claude Desktop and Cursor AI, while helping you quickly find and reuse past work.

How to use

You interact with the Work Memory MCP Server through an MCP client that supports JSON-RPC style requests. Start or switch sessions by project, then add memories, manage todos, and perform searches to retrieve relevant context. The server links memories to sessions so you can maintain a consistent workspace across tools. When you identify information as important, you can assign an importance score and let the system organize and prioritize it. Use the memory, search, session, history, and system tools to create a cohesive, searchable knowledge base for your projects.

How to install

git clone https://github.com/your-repo/work-memory-mcp.git
cd work-memory-mcp
npm install
npm run build
npm start

Configuration and usage tips

Configure client access and environment for the Work Memory MCP Server. The server runs locally via Node and is intended to be invoked by an MCP client. The following configuration snippet shows how to start the server process from a client configuration.

{
  "mcpServers": {
    "work_memory": {
      "command": "node",
      "args": ["/PATH/work-memory/dist/index.js"],
      "env": {
        "WORK_MEMORY_DIR": "/PATH/work-memory/data/",
        "LOG_LEVEL": "WARN",
        "NODE_ENV": "production"
      }
    }
  }
}

Security and data safety

The server uses a local SQLite database to keep memories on your machine and includes input validation to prevent SQL injection. It supports automatic backups and safe cleanup to minimize risk when removing outdated or low-priority data.

Troubleshooting

If the MCP server has connection or performance issues, try restarting the server and verifying your configuration paths.

# Restart the server after a clean build
npm run build
npm start

Available tools

Memory

Core memory management tool for creating, updating, querying, and deleting memory records with tags, project association, and priority.

Search

Full-text search and keyword analysis to locate memories and associated data efficiently.

Session

Session management for project-based workspaces with activation, deactivation, and automatic detection.

History

Versioning and history tracking for memories with restore and compare capabilities.

System

System monitoring and optimization tools for the MCP server, including maintenance and cleanup tasks.