Mnemon Mcp is a local-first memory server for AI agents implementing the Model Context Protocol (MCP). It provides structured, persistent, layered memory for any MCP-compatible client—such as OpenClaw, Claude Code, Cursor, or custom agents—using a simple SQLite database file on the user's machine. Key features include layered memory (episodic, semantic, procedural, resource), fact versioning, hybrid/vector text search, batch import of content, and HTTP or CLI operation, all without cloud or telemetry requirements. Developers and AI practitioners use it to give their AI agents long-term, queryable memory without external dependencies.
Visit Mnemon Mcp's official website for product details and getting started.
Comprehensive user guide and API reference for Mnemon Mcp.
Explore various use cases and examples of implementing Mnemon Mcp.