Memory

关于

Memory is a pure Rust, 'all-in-one' Model Context Protocol (MCP) server designed to give AI agents long-term memory capabilities. Unlike complex setups, it's a single, self-contained executable that embeds SurrealDB, runs embedding models locally via ONNX, and requires zero external dependencies or API keys. It integrates vector search for semantic similarity, a knowledge graph for entity relationships, and code indexing for codebase understanding, using hybrid retrieval for optimal results.

主要功能

• Single, self-contained executable with embedded database (SurrealDB)
• 0 GitHub stars
• Local embedding models (FastEmbed) run on CPU, no API keys required
• Combines vector search, knowledge graph, and code intelligence
• Implements agent protocols for context window compaction and session restarts
• Exposes 26 tools for memory management, search, graph interaction, and codebase intelligence

使用案例

• Integrating long-term memory with AI IDEs like Claude Desktop, Cursor, and OpenCode
• Enabling AI agents to maintain context across sessions and context window resets
• Providing AI agents with semantic understanding of codebases and entity relationships