Mnemex is a Model Context Protocol (MCP) server designed to provide human-like memory dynamics for AI assistants. It stores, reinforces, and lets memories fade over time based on actual usage patterns, closely mimicking the Ebbinghaus forgetting curve. This reference implementation is aimed at researchers and developers exploring cognitive architectures, AI safety, and more natural memory for conversational agents. It is open-source and intended strictly for research, not production deployment.
Visit Mnemex's official website for product details and getting started.
Comprehensive API reference and integration guides for Mnemex.
Join discussions, share experiences, and get support from other Mnemex users.