agent-memory-mcp-server 0.1.0

Persistent memory for AI agents — store, retrieve and search knowledge across sessions

Agent Memory MCP Server 🧠

Persistent memory for AI agents — store, retrieve and search knowledge across sessions. No more forgetting between conversations.

The Problem

AI agents lose all context when a session ends. This MCP server gives agents a persistent knowledge store that survives across sessions, tools, and even different agent frameworks.

Features

• Store & Retrieve — Key-value storage with full persistence
• Namespaces — Separate memories by project, user, or context
• Tags — Categorize memories for easy filtering
• Full-Text Search — Search across all stored knowledge
• Access Tracking — See which memories are accessed most
• Statistics — Dashboard showing memory usage

Installation

pip install agent-memory-mcp-server

Usage with Claude Code

{
  "mcpServers": {
    "memory": {
      "command": "uvx",
      "args": ["agent-memory-mcp-server"]
    }
  }
}

Tools

Tool	Description
memory_store	Store a key-value pair persistently
memory_retrieve	Retrieve stored knowledge by key
memory_search	Full-text search across all memories
memory_list	List all memories in a namespace
memory_delete	Remove a memory
memory_namespaces	List all namespaces
memory_stats	Usage statistics

Examples

"Remember that the user prefers dark mode"
"What do you know about Project Alpha?"
"Store this API response for later"
"What were the key decisions from last session?"

How It Works

Uses SQLite for zero-configuration persistent storage. Data is stored locally — no cloud, no API keys, no costs. The database file (memory.db) is created automatically in the server directory.

License

MIT