MCP Brain Tools

🚀 MCP Memory

MCP Memory is a tool designed to provide persistent memory for AI systems. It supports the creation, management, and search of knowledge graphs through a command-line interface. With MCP Memory, AI systems can remember information from previous conversations and reference it in subsequent exchanges.

🚀 Quick Start

To quickly start using MCP Memory, follow these steps:

1. Clone the MCP repository:

   git clone https://github.com/mcp-servers/mcp-servers.git
   

2. Install dependencies:

   cd mcp-servers && npm install
   

3. Start the MCP Memory service:

   node dist/index.js
   

4. Interact with MCP Memory using the command-line tool:

   # Search for information in memory
   node dist/admin-cli.js search "Project Deadline"
   
   # View detailed information about a specific entity
   node dist/admin-cli.js entity "Project Name"
   
   # Back up the entire memory system
   node dist/admin-cli.js backup memory-backup.json
   

✨ Features

Intelligent Entity Management

MCP Memory offers intelligent entity creation and update capabilities:

• When attempting to create an existing entity, the system returns the existing entity data and guides you on how to add new information.
• The system can distinguish between creating a new entity and updating an existing one.
• Entity relationships are automatically maintained during information updates.

Advanced Features

Memory Zones

You can organize knowledge into different domains:

# Create a project-specific memory zone
node dist/admin-cli.js zones add client-acme "Knowledge about ACME Company Projects"

# Import data into the zone
node dist/admin-cli.js import acme-data.json client-acme

Conversation Memory Management

You can organize memories into different domains through natural conversation instructions to the assistant:

Create and Use Memory Zones

User: Claude, can you create a new memory zone named "Work Projects" to store information about my current work?

Claude: I've created a new memory zone named "Work Projects". I'll store information related to your work in this zone. What information about the project would you like me to remember?

User: The project deadline is March 15th, and we're building a new inventory management system.

Claude: I've stored the information you mentioned in the "Work Projects" zone, including the March 15th deadline and the fact that we're building a new inventory management system. I'll organize any future information related to this project in this zone.
(Behind the scenes: Claude creates a new memory zone and stores project information specifically there)

Retrieve Information from Specific Domains

User: What do you remember about my work project?

Claude: Based on our previous discussion about your work, I remember that you're building an inventory management system for a deadline on March 15th.
(Behind the scenes: Claude specifically searches the "Work Projects" zone for relevant information)

Switch Between Memory Zones

User: Claude, let's switch to my "Cooking" memory zone now. I want to discuss a recipe.

Claude: I've switched to your "Cooking" memory zone. Which dish would you like to discuss today?

User: What was that pasta recipe we mentioned last month?

Claude: Based on our previous conversation in the "Cooking" zone, you shared a Carbonara pasta recipe that includes bacon, eggs, Pecorino Romano cheese, and black pepper. You mentioned it's a family favorite and you learned it during your trip to Rome.
(Behind the scenes: Claude updates its memory context to prioritize searching in the "Cooking" zone)

By organizing memories into different domains, conversations become more relevant and focused on the current topic or project.

Search Function

Leverage the powerful search capabilities of Elasticsearch:

# Fuzzy search (find misspellings of "meeting")
node dist/admin-cli.js search "met"