Memory

About

This tool is a high-performance Rust implementation of the Model Context Protocol (MCP) Memory Server, designed to serve as a robust knowledge graph for persistent AI memory. It acts as a drop-in replacement for the original TypeScript version, offering significant architectural and performance improvements. Leveraging a SQLite backend, it ensures ACID-compliant data integrity, efficient full-text search capabilities with FTS5, and automatic deduplication and cascade deletes. This server is ideal for AI applications requiring reliable, scalable, and fast memory management for complex knowledge structures.

Key Features

• Full-Text Search (FTS5) for efficient searching across graph elements
• 1 GitHub stars
• O(log n) Indexed Queries for rapid lookups, inserts, and deletes
• Async I/O (Tokio-based) for non-blocking operations and high performance
• Automatic Deduplication and Cascade Deletes ensuring data integrity
• SQLite Backend for fast, reliable, ACID-compliant storage

Use Cases

• Debugging and testing AI memory interactions within a robust server environment
• Enabling remote access and integration with web applications for AI memory management
• Providing persistent memory for local AI clients like Claude Desktop, Claude Code, and Codex