Collected molecules will appear here. Add from search or explore.
An MCP (Model Context Protocol) server that provides automated long-term memory for the Claude Code CLI using prompt-intercept hooks to store and retrieve context.
Defensibility
stars
2
The 'claude-memory-manager' is a very early-stage (4 days old, 2 stars) implementation of a memory system for Anthropic's Claude Code CLI. While it intelligently utilizes the 'UserPromptSubmit' hook to automate memory storage—a clever use of the MCP architecture—it suffers from high frontier risk. Anthropic is actively developing Claude Code and the MCP ecosystem; native persistent memory is a core feature likely to be integrated directly into the platform or provided via an official MCP server (similar to their existing 'memory' MCP reference). The project currently lacks a significant user base, unique data moat, or complex algorithmic core that would prevent it from being Sherlocked by a minor update to Claude Code. In the competitive landscape, it sits alongside projects like Mem0 or the official MCP-server-memory, but with a narrower focus on the Claude Code lifecycle. Its defensibility is currently minimal as it functions more as a utility wrapper for existing APIs than a standalone infrastructure project.
TECH STACK
INTEGRATION
cli_tool
READINESS