Shodh Memory

About Shodh Memory

Shodh Memory is a neuroscience-inspired persistent cognitive memory system designed for AI agents and autonomous systems. It features Hebbian learning, knowledge graphs, and a hybrid decay model, enabling memories to strengthen with use and decay over time. Lightweight (single ~30MB binary), fully offline, and edge-first, it's suitable for robots, air-gapped networks, IoT, and agents needing reliable, privacy-preserving long-term memory. It also features Model Context Protocol (MCP) integration for modern agent interoperability.