memory-bank-loader-mcp

Memory Bank Loader MCP Server

memory-bank-loader-mcp gathers files from a memory-bank directory into a single read operation for AI project-context workflows. It recursively collects files and applies a defined ordering, aiming to reduce repeated tool calls when loading context.

Server facts

memory-bank-loader-mcp

Tech

Runtime / language
node
Transports
stdio
Remote available
No
Authentication
Not specified

Activity

GitHub stars
4
Last commit
Data as of

Overview

Loading a project's memory bank is the central workflow for memory-bank-loader-mcp: it gathers directory contents together rather than requiring separate reads for individual files. Its read_memory_bank tool takes a required absolute directory path, reads files recursively, and returns the list of files read. This supports AI workflows that use a Memory Bank as project context. The stated aim is to reduce how often tools must be invoked during context loading, with time and cost savings as the motivation rather than a quantified outcome.

File ordering is part of that context preparation. Files at the directory root take precedence, with designated documents such as projectbrief.md receiving priority there. Other files are ordered alphabetically using their relative paths, including those found in nested directories. The project runs on Node and communicates over stdio. Its setup examples cover launching through NPX or running the built JavaScript entry point with Node, and the code is released under the MIT license.

memory-bank-loader-mcp setup notes

  • memory-bank-loader-mcp records stdio only; confirm a network transport upstream before planning remote access.

Related MCP servers