MCP Catalogs
HomeMemory-Plus screenshot

Memory-Plus

by Yuchen20·58·Score 43

Lightweight local RAG memory store for MCP agents with record, retrieve, update, and visualization features.

ai-llmdeveloper-toolsknowledge-graph
7
Forks
2
Open issues
12 mo ago
Last commit
2d ago
Indexed

Overview

Memory-Plus is a Python-based MCP server that provides a local Retrieval-Augmented Generation (RAG) memory store for AI agents. It allows developers to record, retrieve, update, delete, and visualize persistent memories across sessions. The server uses Google's Gemini Embedding API for memory processing and maintains versions of memories for historical context. It's particularly useful for developers working with multiple AI coding tools who want their AI assistants to maintain context between sessions.

Try asking AI

After installing, here are 5 things you can ask your AI assistant:

you:Enable AI coding assistants to remember project context across sessions
you:Store and retrieve user preferences for personalized AI interactions
you:Create knowledge graphs of ideas and concepts for better understanding
you:Why is memory-plus not working?
you:How do I use memory-plus in a real chat session?

When to choose this

Choose Memory-Plus when you need a local, privacy-focused memory solution for your AI agents that works across multiple sessions and IDEs without requiring external services.

When NOT to choose this

Don't choose Memory-Plus if you need cloud-based memory sharing across devices, require advanced search capabilities beyond basic keyword matching, or want a solution without external API dependencies like Google's.

Tools this server exposes

7 tools extracted from the README
  • record_memory

    Save user data, ideas, and important context.

  • retrieve_memory

    Search by keywords or topics over past entries.

  • get_recent_memories

    Fetch the last N memory items.

  • update_memory

    Append or modify existing memory entries.

  • visualize_memories

    Create interactive graph clusters revealing relationships between memories.

  • import_file

    Ingest documents directly into memory.

  • delete_memory

    Remove unwanted memory entries.

Comparable tools

semantic-memory-mcpmemgptautogen-mcpvector-mcp

Installation

Installation

**Prerequisites:**

  1. Google API Key (for Gemini Embedding API)

- Obtain from [Google AI Studio](https://aistudio.google.com/apikey) - Set as GOOGLE_API_KEY in environment

**Setup with UV Runtime:**

# Install UV
pip install uv

# Add to VS Code/settings.json
{
  "mcpServers": {
    "memory-plus": {
      "command": "uvx",
      "args": [
        "-q",
        "memory-plus@latest"
      ],
      "env": {
        "GOOGLE_API_KEY": "<YOUR_API_KEY>"
      }
    }
  }
}

**One-Click VS Code Install:** [![One Click Install in VS Code](https://imgix.net/memory-plus.png)](https://insiders.vscode.dev/redirect/mcp/install?name=memory-plus&config=%7B%22command%22%3A%22uvx%22%2C%22args%22%3A%5B%22-q%22%2C%22memory-plus%40latest%22%5D%7D)

FAQ

Why is memory-plus not working?
Memory-plus has dependencies that can be slow to download the first time (about 1 minute). Once installed, subsequent usage is much faster.
How do I use memory-plus in a real chat session?
Add the MCP JSON configuration to your MCP setup. Once added, memory-plus will automatically activate when needed during conversations.

Compare Memory-Plus with

GitHub →

Last updated · Auto-generated from public README + GitHub signals.