Memory Bank MCP Server 2.2.1

A memory bank server based on the Model Context Protocol (MCP) for cross - session management of project documents and context, helping AI agents (such as Claude) maintain consistent project knowledge through structured JSON - formatted global and branch - specific memory banks.

Knowledge management and memory Developer tools #Memory Management #MCP Protocol #AI Assistance #Document Storage .TypeScript

rating : 2.5 points

downloads : 39

update time : 2025-04-23

What is Memory Bank MCP Server?

This is an intelligent document management system specifically designed for AI agents (such as Claude). It uses the Model Context Protocol (MCP) to help AI remember project - related documents and context information, achieving cross - session knowledge consistency.

How to use Memory Bank MCP Server?

Simply specify your project document directory, and the server will automatically create and manage the global memory bank (global - memory - bank) and branch memory bank (branch - memory - bank). AI agents can access this structured knowledge through the MCP protocol.

Use Cases

It is particularly suitable for long - term development projects, collaborative scenarios that require maintaining context consistency across multiple AI sessions, and complex tasks that require AI to continuously learn project knowledge.

Main Features

Intelligent Memory ManagementAutomatically maintain global and branch - level memory banks to ensure that AI agents always access the latest project knowledge

MCP Protocol SupportAchieve efficient communication with AI agents through the standardized Model Context Protocol

VSCode IntegrationProvide a VSCode extension to facilitate developers to directly manage the memory bank in the IDE

Structured StorageUse JSON Schema to standardize the format of memory documents, ensuring data consistency and queryability

Advantages and Limitations

Advantages

Maintain long - term memory consistency of AI agents

Reduce the time for repeatedly explaining project background

Support multi - branch parallel development scenarios

Open - source and easy to expand

Limitations

Require initial setup of the document directory structure

Currently mainly support text - based knowledge management

Require a Node.js environment to run

How to Use

Installation Preparation

Ensure that Node.js and Yarn (v1.x) are installed

Clone and Install

Get the project code and install dependencies

Start the Server

Specify your project document path to start the MCP server

Use the Memory Bank

AI agents can now access the knowledge in the memory bank through the MCP protocol

Usage Examples

Long - term Project DevelopmentIn a 3 - month project, the AI agent maintains continuous memory of the project architecture and API specifications through the memory bank

Multi - branch CollaborationThe team is working on the feature/login and feature/payment branches simultaneously. The AI can provide accurate context based on the current branch

Frequently Asked Questions

How is the data in the memory bank stored?

Does it support integration with other IDEs?

How to update the content in the memory bank?

Related Resources

GitHub Repository

Project source code and latest updates

MCP Protocol Documentation

Detailed technical specifications of the Model Context Protocol

Example Project

A reference implementation for inspiration

🚀 Memory Bank MCP Server

A memory bank implementation for managing project documents and technical backgrounds to maintain consistency across different sessions. This server uses the Model Context Protocol (MCP) to help AI agents (such as Claude) maintain knowledge consistency across projects. This knowledge is stored in the form of global and branch-specific memory banks in a structured JSON format.

This project is inspired by Cline Memory Bank from the nickbaumann98/cline_docs repository.

✨ Features

Packages

This repository is a monorepo managed by Yarn workspaces. It includes the following packages:

packages/mcp: The core implementation of the MCP server. It contains the main logic for handling memory bank operations, MCP tool execution, and server startup.
packages/schemas: Defines JSON schemas for memory bank documents (e.g., memory_document_v2).
packages/vscode-extension: A VSCode extension that provides integration with the Memory Bank MCP server (details to be determined).

🚀 Quick Start

Prerequisites

Node.js (Refer to .tool-versions to ensure you are using the recommended version).
Yarn (v1.x)

📦 Installation

Clone the repository and install dependencies from the root directory:

git clone https://github.com/t3ta/memory-bank-mcp-server.git
cd memory-bank-mcp-server
yarn install

Running the MCP Server

You can run the MCP server directly from the monorepo:

# Execute at the monorepo root directory
yarn workspace @memory-bank/mcp start --docs /path/to/your/docs

Replace /path/to/your/docs with the actual path to your project documents (where the global-memory-bank and branch-memory-bank files or folders will be created or located).

For more information on running the server and available options, refer to packages/mcp/README.md.