MCP Catalogs
Home

kept vs memory

Side-by-side comparison to help you pick between these two MCP servers.

kept
by egroup-labs
memory
by modelcontextprotocol
Stars★ 83★ 85,748
30d uses
Score4877
Official
Categories
File SystemKnowledge GraphProductivity
Knowledge GraphAI / LLM ToolsProductivity
LanguageTypeScriptTypeScript
Last committhis monththis month

kept · Summary

Kept is an MCP server that manages AI conversations as searchable Markdown files in a local vault.

memory · Summary

An MCP server implementing persistent memory using a local knowledge graph for AI models to remember user information across chats.

kept · Use cases

  • Search and reuse past AI conversations within coding workflows
  • Maintain a personal knowledge graph of AI interactions
  • Archive and organize research notes from AI assistants

memory · Use cases

  • Personalizing AI assistant interactions by remembering user preferences, history, and relationships
  • Building context-aware chat applications that maintain conversation history
  • Creating knowledge bases that persist across AI model sessions

kept · Install

Installation

Desktop App

Quick install:

# macOS/Linux
curl -fsSL https://kept.work/install.sh | bash

# Windows PowerShell
irm https://github.com/egroup-labs/kept/releases/latest/download/install.ps1 | iex

MCP Server

# macOS/Linux
bash <(curl -fsSL https://raw.githubusercontent.com/egroup-labs/kept/main/scripts/install-kept-mcp.sh)

# Windows PowerShell
& ([scriptblock]::Create((irm https://raw.githubusercontent.com/egroup-labs/kept/main/scripts/install-kept-mcp.ps1)))

Claude Desktop

Add to claude_desktop_config.json:

{
  "mcpServers": {
    "kept-vault": {
      "command": "npx",
      "args": ["@kept-plugins/kept-vault"]
    }
  }
}

Set KEPT_VAULT_PATH environment variable if your vault is not at ~/.kept/vault.

memory · Install

Installation

Claude Desktop

Add to your claude_desktop_config.json:

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

VS Code

Use one-click installation buttons or manually configure in .vscode/mcp.json:

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

Docker

{
  "mcpServers": {
    "memory": {
      "command": "docker",
      "args": ["run", "-i", "-v", "claude-memory:/app/dist", "--rm", "mcp/memory"]
    }
  }
}
Comparison generated from public README + GitHub signals. Last updated automatically.