Skip to main content
Glama

MCP Memory Server

by ebailey78
# Changelog ## v0.1.0 - Initial Release ### Features - Created a Model Context Protocol (MCP) server for Claude memory management - Implemented memory storage using markdown files - Added Lunr.js for memory indexing and search - Implemented memory types: entity, concept, and session - Added tagging and relationship capabilities - Created a build_memory_store tool for initializing memory stores in specified directories - Added comprehensive instructions template for Claude projects - Designed for project-specific memory management within project directories ### Technical Improvements - Used the official MCP SDK for compatibility with Claude Desktop - Implemented proper type safety with TypeScript and Zod - Added error handling for all operations - Created test scripts to verify functionality ### Documentation - Added README.md with usage instructions - Created USAGE.md with detailed tool documentation - Added instructions_template.md for Claude project instructions - Created example configuration files for Claude Desktop - Emphasized project-specific workflow and memory organization ### Next Steps - Add more comprehensive tests - Implement vector embeddings for semantic search - Add memory prioritization based on importance and recency - Create a web interface for memory visualization - Add automatic tagging based on content analysis

MCP directory API

We provide all the information about MCP servers via our MCP API.

curl -X GET 'https://glama.ai/api/mcp/v1/servers/ebailey78/mcp-memory'

If you have feedback or need assistance with the MCP directory API, please join our Discord server