Memory Plus
π§ π΄πππππ-π·πππ is a lightweight, local RAG memory store for MCP agents. Easily record, retrieve, update, delete, and visualize persistent 'memories' across sessionsβperfect for developers working with multiple AI coders (like Windsurf, Cursor, or Copi
What is Memory Plus?
Memory Plus is a Model Context Protocol (MCP) server that allows AI assistants like Claude, Cursor, and VS Code to π§ π΄πππππ-π·πππ is a lightweight, local rag memory store for mcp agents. easily record, retrieve, update, delete, and visualize persistent 'memories' across sessionsβperfect for developers wo...
π§ π΄πππππ-π·πππ is a lightweight, local RAG memory store for MCP agents. Easily record, retrieve, update, delete, and visualize persistent 'memories' across sessionsβperfect for developers working with multiple AI coders (like Windsurf, Cursor, or Copi
This server falls under the Knowledge & Memory category on MCPgee, the world's largest MCP server directory with 33,000+ servers.
Features
- π§ π΄πππππ-π·πππ is a lightweight, local RAG memory store for M
Use Cases
Maintainer
Works with
Installation
NPM
npx -y memory-plusManual Installation
npx -y memory-plusConfiguration
Configuration Details
claude_desktop_config.json
Performance
Response Metrics
Resource Usage
How to Set Up and Use Memory Plus
Memory Plus is a lightweight local RAG (Retrieval-Augmented Generation) memory store for MCP agents that lets AI assistants record, retrieve, update, and delete persistent memories across sessions using Google's Gemini Embedding API for semantic similarity. It stores all memories locally as files (no cloud database required) and includes an interactive graph visualisation for exploring the memory network. Developers using Windsurf, Cursor, Claude Code, or Copilot use it to give their AI coding assistants persistent project knowledge that survives context resets.
Prerequisites
- Python 3.10 or later and uv installed (recommended) or pip
- A Google AI Studio API key for the Gemini Embedding API (free tier available at aistudio.google.com/apikey)
- An MCP-compatible client such as Claude Code, Cursor, or Windsurf
- The GOOGLE_API_KEY environment variable set to your API key
Obtain a Google AI Studio API key
Visit aistudio.google.com/apikey, sign in with a Google account, and generate a new API key. Copy itβit is used by Memory Plus for embedding generation.
Install uv (recommended runner)
uv provides dependency-isolated execution via uvx. Install it with the official installer or via pip.
curl -LsSf https://astral.sh/uv/install.sh | shOpen your MCP client configuration
For Claude Code edit ~/.claude/settings.json or the project-level .mcp.json. For Cursor open its MCP settings file (usually .cursor/mcp.json).
Add the Memory Plus server entry
Add the mcpServers block with your GOOGLE_API_KEY. Using uvx ensures the latest version of memory-plus is used without a manual install step.
{
"mcpServers": {
"memory-plus": {
"command": "uvx",
"args": ["-q", "memory-plus@latest"],
"env": {
"GOOGLE_API_KEY": "your-google-api-key"
}
}
}
}Restart your MCP client
Reload Claude Code or Cursor to start the Memory Plus server. uvx will download the package on first run.
Store your first memory
Ask your AI assistant to record a memory. From this point on, the agent can recall relevant memories automatically at the start of new sessions.
Memory Plus Examples
Client configuration
MCP config for Claude Code using uvx. Replace the GOOGLE_API_KEY value with your actual key from Google AI Studio.
{
"mcpServers": {
"memory-plus": {
"command": "uvx",
"args": ["-q", "memory-plus@latest"],
"env": {
"GOOGLE_API_KEY": "your-google-api-key"
}
}
}
}Prompts to try
Example requests that exercise the memory record, retrieve, search, update, and visualise capabilities.
- "Remember that we use PostgreSQL 15 with the pgvector extension for this project"
- "What do you remember about the database setup for this project?"
- "Search your memories for anything related to authentication or JWT tokens"
- "Show me the most recent memories you have stored"
- "Update your memory about the API rate limits to reflect the new 1000 req/min limit"
- "Delete all memories tagged with the 'temporary' keyword"
- "Open the memory graph visualisation so I can see how my memories are connected"Troubleshooting Memory Plus
API key error: GOOGLE_API_KEY is invalid or missing
Ensure the GOOGLE_API_KEY env var is set in your MCP config and that the key is active at aistudio.google.com/apikey. The free tier includes sufficient quota for typical memory operations.
uvx command not found
uv is not installed or not in your PATH. Install it with: curl -LsSf https://astral.sh/uv/install.sh | sh (macOS/Linux). After installation, restart your terminal or run 'source ~/.zshrc'. Alternatively install via pip: pip install uv.
Memories are not persisted between sessions
Memory Plus stores data in a local directory. Ensure the process has write access to its working directory. If you run different MCP clients under different user accounts, each will have a separate memory store. Check the logs for any file permission errors.
Frequently Asked Questions about Memory Plus
What is Memory Plus?
Memory Plus is a Model Context Protocol (MCP) server that π§ π΄πππππ-π·πππ is a lightweight, local rag memory store for mcp agents. easily record, retrieve, update, delete, and visualize persistent 'memories' across sessionsβperfect for developers working with multiple ai coders (like windsurf, cursor, or copi It connects AI assistants to external tools and data sources through a standardized interface.
How do I install Memory Plus?
Install via npm with the command: npx -y memory-plus. Then add the server configuration to your AI client's JSON config file (e.g., claude_desktop_config.json or .cursor/mcp.json).
Which AI clients work with Memory Plus?
Memory Plus works with all major MCP-compatible AI clients including Claude Desktop, Claude Code, Cursor, VS Code (GitHub Copilot), Windsurf, and Cline.
Is Memory Plus free to use?
Yes, Memory Plus is open source and available under the Apache-2.0 license. You can use it freely in both personal and commercial projects.
Memory Plus Alternatives β Similar Knowledge & Memory Servers
Looking for alternatives to Memory Plus? Here are other popular knowledge & memory servers you can use with Claude, Cursor, and VS Code.
MemPalace
β 52.6kA local AI memory system that stores all conversations verbatim and organizes them into navigable structures. It provides 19 MCP tools for AI assistants to search and retrieve past decisions, debugging sessions, and architecture debates automatically
Kratos
β 25.7kποΈ Memory System for AI Coding Tools - Never explain your codebase again. MCP server with perfect project isolation, 95.8% context accuracy, and the Four Pillars Framework.
Context Mode
β 15.4kAn MCP server that preserves LLM context by intercepting large data outputs and returning only concise summaries or relevant sections. It enables efficient sandboxed code execution, file processing, and documentation indexing across multiple programm
Memu
β 13.7kMemory for 24/7 proactive agents like OpenClaw.
MemOS
β 9.3kMemOS (Memory Operating System) is a memory management operating system designed for AI applications. Its goal is: to enable your AI system to have long-term memory like a human, not only remembering what users have said but also actively invoking, u
Everos
β 5.4kBuild, evaluate, and integrate long-term memory for self-evolving agents.
Browse More Knowledge & Memory MCP Servers
Explore all knowledge & memory servers available in the MCPgee directory. Each server includes setup guides for Claude, Cursor, and VS Code.
Set Up Memory Plus in Your Editor
Choose your AI client for step-by-step setup instructions.
Quick Config Preview
Add this to your claude_desktop_config.json or .cursor/mcp.json
Ready to use Memory Plus?
Browse our complete directory of 33,000+ MCP servers, read setup guides for your editor, and start building with the Model Context Protocol.