Memory MCP

v1.0.0Knowledge & Memorystable

MCP server for Ragionex Memory - persistent semantic memory for AI agents. Works in Claude Code, Claude Desktop, Codex, ChatGPT (Apps SDK), Cursor, Cline, and 500+ MCP clients. Save in one tool, recall in another. Project-scoped, cross-session, agent

agentic-aiai-agentschatgptclaudeclaude-code
Share:
104
Stars
0
Downloads
0
Weekly
0/5

What is Memory MCP?

Memory MCP is a Model Context Protocol (MCP) server that allows AI assistants like Claude, Cursor, and VS Code to mcp server for ragionex memory - persistent semantic memory for ai agents. works in claude code, claude desktop, codex, chatgpt (apps sdk), cursor, cline, and 500+ mcp clients. save in one tool, recal...

MCP server for Ragionex Memory - persistent semantic memory for AI agents. Works in Claude Code, Claude Desktop, Codex, ChatGPT (Apps SDK), Cursor, Cline, and 500+ MCP clients. Save in one tool, recall in another. Project-scoped, cross-session, agent

This server falls under the Knowledge & Memory category on MCPgee, the world's largest MCP server directory with 33,000+ servers.

Features

  • create_entities
  • create_relations
  • add_observations
  • delete_entities
  • delete_observations

Use Cases

Semantic agent memory
Cross-session knowledge persistence
Multi-client compatibility
LicenseMIT License
Languageswift
Versionv1.0.0
UpdatedMay 2, 2026
Statushealthy
Maintenanceactive

Works with

ClaudeOpenAIwindowsmacoslinux

Installation

NPM

npx -y @modelcontextprotocol/server-memory

Manual Installation

npx -y @modelcontextprotocol/server-memory

Configuration

Configuration Details

Config File

claude_desktop_config.json

Performance

Response Metrics

Response Time< 200ms
ThroughputMedium

Resource Usage

Memory UsageLow
CPU UsageLow

How to Set Up and Use Memory MCP

Memory MCP (via @modelcontextprotocol/server-memory) provides a knowledge graph-based persistent memory system for AI agents, enabling them to store and retrieve information across sessions. The server exposes tools for creating entities and relations, adding observations, and searching the knowledge graph so that an AI agent can build up a structured understanding of users, projects, and concepts over time. It works across Claude Code, Claude Desktop, Cursor, Cline, and any other MCP-compatible client, making memory available wherever the agent runs.

Prerequisites

  • Node.js 18+ and npx available on your machine
  • An MCP-compatible client such as Claude Desktop, Claude Code, or Cursor
  • No API key required — the server uses local file-based storage by default
  • Optional: sufficient disk space for the knowledge graph storage file (typically small)
1

Run the server directly with npx

The @modelcontextprotocol/server-memory package can be run directly without installation using npx. This is the simplest way to get started.

npx -y @modelcontextprotocol/server-memory
2

Add to Claude Desktop configuration

Edit your Claude Desktop config file to register the memory server so it starts automatically with Claude Desktop.

3

Restart Claude Desktop

After editing the configuration file, restart Claude Desktop completely to load the memory server.

4

Verify the memory tools are available

Ask Claude to use the memory tools to store a piece of information, confirming the server is connected and working.

5

Test cross-session persistence

Store a fact in one session, then start a new Claude Desktop session and ask it to recall that fact to verify cross-session memory is working.

Memory MCP Examples

Client configuration

Claude Desktop configuration for the official @modelcontextprotocol/server-memory package. The server uses local file storage with no additional environment variables required.

{
  "mcpServers": {
    "memory": {
      "command": "npx",
      "args": ["-y", "@modelcontextprotocol/server-memory"]
    }
  }
}

Prompts to try

Example prompts to use the memory server for storing and retrieving knowledge across sessions.

- "Remember that my preferred coding style is 2-space indentation with TypeScript strict mode."
- "Create an entity for my project 'ProjectAlpha' and note that it uses React 18 and a PostgreSQL database."
- "What do you know about my project ProjectAlpha?"
- "Add an observation that I fixed the authentication bug on 2026-06-10."
- "Search your memory for everything you know about my database setup."

Troubleshooting Memory MCP

Memory does not persist between Claude Desktop sessions

Ensure the memory server is listed in your claude_desktop_config.json and that Claude Desktop is restarted after editing the config. The knowledge graph is stored in a local file; if you are running the server ad-hoc with npx rather than via the config, each run may use a different working directory and thus a different storage file.

npx -y @modelcontextprotocol/server-memory fails with a package not found error

Verify your npm/npx installation with `npx --version`. If outdated, update npm with `npm install -g npm@latest`. The package is published on the official npm registry; ensure you are not behind a corporate proxy that blocks npm registry access.

Memory tools appear but create_entities or add_observations returns an error

Check that the server process has write access to its working directory where the knowledge graph file is stored. On macOS, if Claude Desktop is sandboxed, try specifying an explicit writable path for storage by consulting the server's documentation for storage path configuration options.

Frequently Asked Questions about Memory MCP

What is Memory MCP?

Memory MCP is a Model Context Protocol (MCP) server that mcp server for ragionex memory - persistent semantic memory for ai agents. works in claude code, claude desktop, codex, chatgpt (apps sdk), cursor, cline, and 500+ mcp clients. save in one tool, recall in another. project-scoped, cross-session, agent It connects AI assistants to external tools and data sources through a standardized interface.

How do I install Memory MCP?

Install via npm with the command: npx -y @modelcontextprotocol/server-memory. Then add the server configuration to your AI client's JSON config file (e.g., claude_desktop_config.json or .cursor/mcp.json).

Which AI clients work with Memory MCP?

Memory MCP works with all major MCP-compatible AI clients including Claude Desktop, Claude Code, Cursor, VS Code (GitHub Copilot), Windsurf, and Cline.

Is Memory MCP free to use?

Yes, Memory MCP is open source and available under the MIT License license. You can use it freely in both personal and commercial projects.

Browse More Knowledge & Memory MCP Servers

Explore all knowledge & memory servers available in the MCPgee directory. Each server includes setup guides for Claude, Cursor, and VS Code.

Quick Config Preview

{ "mcpServers": { "memory": { "command": "npx", "args": ["-y", "@modelcontextprotocol/server-memory"] } } }

Add this to your claude_desktop_config.json or .cursor/mcp.json

Read the full setup guide →

Ready to use Memory MCP?

Browse our complete directory of 33,000+ MCP servers, read setup guides for your editor, and start building with the Model Context Protocol.

33,000+ ServersFree & Open SourceStep-by-Step Guides