mcp-openmemory

Overview

MCP OpenMemory is a standalone server that provides Claude with persistent conversation memory capabilities. It stores conversation messages, maintains summarized memory contexts, and provides access to recent conversations within configurable time windows. The server uses SQLite for local storage without requiring external dependencies, making it privacy-focused and self-contained.

Try asking AI

After installing, here are 5 things you can ask your AI assistant:

• save_memory

  Store individual conversation messages

  Try:Save this important conversation detailTry:Remember this key information about my preferences
• recall_memory_abstract

  Get current memory summary

  Try:What do you remember about our previous conversation?Try:Recall my memory abstract
• update_memory_abstract

  Update the memory summary

  Try:Update my memory with this new project informationTry:Refresh my memory summary with the latest developments
• get_recent_memories

  Retrieve recent conversation history

  Try:Show me my recent conversationsTry:What did we discuss last time?

Comparable tools

Installation

{
  "mcpServers": {
    "mcp-openmemory": {
      "command": "npx",
      "args": [
        "@peakmojo/mcp-openmemory@latest"
      ],
      "env": {
        "MEMORY_DB_PATH": "/path/to/your/persistent/memory.sqlite"
      }
    }
  }
}

FAQ

Where is the memory data stored?

Data is stored in an SQLite database file at the path specified in MEMORY_DB_PATH environment variable. Default is ./memory.sqlite but should be set to a persistent location.

How can I separate memory for different projects?

You can either use different MEMORY_DB_PATH values for each application (hard separation) or pass a custom 'context' parameter when calling memory tools (soft separation).

Compare mcp-openmemory with