memory-agent-mcp

AnsKM/memory-agent-mcp

3.2

If you are the rightful owner of memory-agent-mcp and would like to certify it and/or have it hosted online, please leave a comment on the right or send an email to henry@mcphub.com.

The Memory Agent MCP Server is designed to enhance Claude Code by providing persistent memory capabilities, overcoming context window limitations through efficient information storage and retrieval.

Memory Agent MCP Server

An MCP (Model Context Protocol) server that provides persistent memory capabilities for Claude Code, solving context window limitations through intelligent information storage and retrieval.

Features

  • 🧠 Persistent Memory: Store and retrieve conversations, code snippets, and project context
  • 🔍 Semantic Search: Find relevant information using vector embeddings
  • 🚀 Fast Retrieval: Sub-100ms response times with optimized vector search
  • 💾 Project Isolation: Separate memory stores for different projects
  • 🔄 Workflow Memory: Remember and replay multi-step processes
  • 📊 Smart Ranking: Multi-factor memory ranking (relevance, recency, importance)

Architecture

Built on:

  • FastAPI: Modern async Python web framework
  • Qdrant: High-performance vector database
  • Voyage AI: State-of-the-art embedding models
  • MCP Protocol: Anthropic's standard for AI tool integration

Quick Start

Prerequisites

  • Python 3.9+
  • Docker (optional, for Qdrant)
  • Voyage AI API key

Installation

  1. Clone the repository:
git clone https://github.com/yourusername/memory-agent-mcp.git
cd memory-agent-mcp
  1. Install dependencies:
pip install -e .
  1. Set up environment variables:
cp .env.example .env
# Edit .env with your API keys
  1. Start Qdrant (using Docker):
docker run -p 6333:6333 qdrant/qdrant
  1. Run the MCP server:
memory-agent-mcp

Integration with Claude Code

Add to your Claude Code configuration:

claude mcp add memory "python /path/to/memory-agent-mcp/src/mcp/server.py"

Usage

Once integrated, you can use memory features in Claude Code:

  • @memory recall "regex patterns" - Find past code patterns
  • @memory save-workflow "deployment" - Save multi-step processes
  • @memory project-context - Get project-specific information

Development

Running Tests

pytest

Code Quality

# Format code
black src/

# Lint
ruff check src/

# Type check
mypy src/

Cost

  • Minimal Setup: ~$12/month (2GB VPS)
  • Recommended: ~$27/month (4GB VPS + backups)
  • Embeddings: Free tier covers most usage

License

MIT License - see LICENSE file for details

Acknowledgments

Built on top of the Memorizz framework.