claude mcp add umblebiber-hmem-mcp -- npx -y hmem-mcp
{
"mcpServers": {
"umblebiber-hmem-mcp": {
"command": "npx",
"args": ["-y", "hmem-mcp"]
}
}
}
{
"mcpServers": {
"umblebiber-hmem-mcp": {
"command": "npx",
"args": ["-y", "hmem-mcp"]
}
}
}
{
"mcpServers": {
"umblebiber-hmem-mcp": {
"command": "npx",
"args": ["-y", "hmem-mcp"]
}
}
}
About This MCP Server
The hmem-mcp server provides a persistent 5-level hierarchical memory for AI agents, utilizing SQLite as its backend and implementing lazy loading for efficient data retrieval. It connects to AI coding assistants through standard input/output interfaces, enabling them to store and retrieve data in a structured manner. Developers would use this server when building applications that require a reliable and organized memory system for their AI agents, such as those using Claude Code or Cursor.
Persistent 5-level hierarchical memory for AI agents. SQLite-backed, lazy-loaded.. This is a Model Context Protocol (MCP) server that extends AI assistants like Claude with database capabilities via the stdio transport.
Package
hmem-mcp on npm
HOW TO USE
Select your AI client above to get the install command. This MCP server uses the stdio transport and is available on npm.
What tools does it provide?
The hmem — Humanlike Memory for AI Agents server extends your AI assistant with database capabilities. Once installed, your AI can use its tools automatically.