# Memory Management

ContextBase is an AI context management server based on the MCP protocol, providing functions such as user authentication, memory storage, and logging. It uses technology stacks such as TypeScript, Fastify, and Prisma, and supports PostgreSQL and Redis databases.

An MCP server implementation designed for the Claude desktop application, providing persistent memory functions for large language models, including a tiered memory architecture, multiple memory types, and semantic search features.

An MCP server implementation designed for the Claude desktop application, providing persistent memory functions, including a tiered memory architecture, multiple memory types, and semantic search features.

The Memory Box MCP Server provides tools for Cline and Claude Desktop to interact with Memory Box instances, supporting the saving, searching, and formatting of semantic memories.

MCP Server for Zep Cloud is a bridge service connecting Claude Desktop and Zep Cloud API, providing user management, memory collection operations, and memory storage functions, and supporting an automatic fallback operation mode.

MCP-Mem0 is a Python-based server project focused on providing long-term memory management functions for intelligent agents. It offers efficient memory storage and retrieval capabilities and serves as a template to facilitate developers in building their own MCP servers. The project is lightweight and easy to deploy, suitable for scenarios requiring agent memory management.

A service that accesses Apple device data through the MCP protocol, supporting memory storage and retrieval

This project adds custom memory paths and timestamp functions to the MCP team's Memory server, supporting the construction of a knowledge graph through LLM to manage interaction memory.

An enhanced MCP memory server with functions of dynamic compression, context management, and integration of thinking processes

A memory bank server based on the Model Context Protocol (MCP) for cross - session management of project documents and context, helping AI agents (such as Claude) maintain consistent project knowledge through structured JSON - formatted global and branch - specific memory banks.

A knowledge graph memory server implemented in Go, providing persistent memory management capabilities across conversations for AI assistants

An MCP Memory Server designed for Claude Desktop, providing structured memory management across chat sessions, focusing on project work.

An MCP service for semantic search and memory management based on txtai, providing context memory capabilities for AI assistants

The Memory Custom project adds custom memory paths and timestamp functions to the Memory server of the MCP team, optimizing the management of knowledge graphs and interaction tracking.

This project adds knowledge graph management functionality to the MCP memory server, using the PouchDB database for efficient storage, supporting custom storage paths and timestamp recording.

A server based on the MCP protocol for managing memory texts in Claude chats, supporting operations such as addition, deletion, search, and modification.

This project is a simplified implementation based on the MemGPT research paper, built using Spring Boot and Spring AI, providing two integration methods: REST services and MCP servers, and including core agent functions and chatbot applications.

Unibase membase is a decentralized memory layer designed for AI agents, achieving seamless integration with the decentralized storage protocol through the membase mcp server and supporting the upload and download of memory data.

The core implementation directory of the Aegis Framework, including memory management, template system, operation mode, and task lifecycle management

An MCP server for efficiently processing Figma files, supporting chunking and pagination functions