Shodh-Memory
Shodh-Memory is a cognitive memory system for AI agents. It gives coding assistants persistent memory that strengthens with use (Hebbian learning), decays naturally (Ebbinghaus forgetting curves), and surfaces related context automatically (spreading activation in a knowledge graph). For developers using Cursor, this means: - Your coding assistant remembers your project architecture, preferences, and past decisions across sessions - Frequently-used context gets stronger — it never forgets your tech stack or coding patterns - Related memories surface proactively when you start a new task - Full GTD task management (todos, projects) persists across conversations - 45 MCP tools for memory, recall, todos, reminders, knowledge graph, and more Single binary, runs 100% offline on localhost. No cloud dependency. Apache 2.0 licensed. Published on: [npm](https://www.npmjs.com/package/@shodh/memory-mcp) · [crates.io](https://crates.io/crates/shodh-memory) · [PyPI](https://pypi.org/project/shodh-memory/) · [MCP Registry](https://registry.modelcontextprotocol.io)
Shodh-Memory
Shodh-Memory is a cognitive memory system for AI agents. It gives coding assistants persistent memory that strengthens with use (Hebbian learning), decays naturally (Ebbinghaus forgetting curves), and surfaces related context automatically (spreading activation in a knowledge graph). For developers using Cursor, this means: - Your coding assistant remembers your project architecture, preferences, and past decisions across sessions - Frequently-used context gets stronger — it never forgets your tech stack or coding patterns - Related memories surface proactively when you start a new task - Full GTD task management (todos, projects) persists across conversations - 45 MCP tools for memory, recall, todos, reminders, knowledge graph, and more Single binary, runs 100% offline on localhost. No cloud dependency. Apache 2.0 licensed. Published on: [npm](https://www.npmjs.com/package/@shodh/memory-mcp) · [crates.io](https://crates.io/crates/shodh-memory) · [PyPI](https://pypi.org/project/shodh-memory/) · [MCP Registry](https://registry.modelcontextprotocol.io)