
Memory Bank MCP
A Model Context Protocol plugin that helps AI assistants maintain persistent project context through structured markdown files, providing a systematic approach to tracking project goals, decisions, progress, and patterns.
README
Memory Bank MCP
A guided Memory Bank plugin for AI-assisted development
Memory Bank MCP is a Model Context Protocol (MCP) plugin that helps AI assistants maintain persistent project context through structured markdown files. It provides a systematic approach to tracking project goals, decisions, progress, and patterns through guided instructions rather than direct operations.
Features
- Guided Operations: Provides instructions for AI assistants to perform operations themselves
- Structured Context Management: Organize project information across 5 core files
- Intelligent Guidance: Step-by-step instructions for initialization and updates
- Flexible Updates: Smart update guidance based on different change types
- Cross-Platform Support: Automatic path normalization for Windows/macOS/Linux
MCP Configuration
Using the published npm package:
{
"mcpServers": {
"memory-bank": {
"command": "npx",
"args": ["@neko0721/memory-bank-mcp"],
"timeout": 600
}
}
}
Quick Start
-
Initialize Memory Bank
Use init-memory-bank to create the memory-bank directory and core files
-
Read Memory Bank
Use get-memory-bank-info to view all Memory Bank content
-
Update Memory Bank
Use update-memory-bank to get guidance on updating specific files
Core Files
1. productContext.md (Product Context)
- High-level project overview
- Goals and key features
- Overall architecture
- Automatically incorporates projectBrief.md if available
2. activeContext.md (Active Context)
- Current work status
- Recent changes
- Open questions and issues
- Focus areas
3. progress.md (Progress)
- Task tracking in checklist format
- Completed, current, and planned tasks
- Progress timeline
4. decisionLog.md (Decision Log)
- Architectural and implementation decisions
- Rationale and implications
- Decision history
5. systemPatterns.md (System Patterns)
- Recurring patterns and standards
- Coding conventions
- Architectural patterns
- Testing strategies
Usage Guidelines
For AI Assistants
- Start Every Session: Check if memory-bank directory exists, then use
get-memory-bank-info
to understand project state - Initialize When Needed: Use
init-memory-bank
for new projects - Read Context: Use
get-memory-bank-info
to understand project state - Update Guidance: Use
update-memory-bank
to get update instructions - Follow Instructions: Execute the provided guidance to maintain Memory Bank
Update Triggers
- Architecture Changes: Major structural decisions
- Feature Completion: New features or capabilities
- Bug Fixes: Significant issue resolutions
- Refactoring: Code structure improvements
- Decisions: Any important technical choices
- Progress Updates: Task status changes
Tool Reference
init-memory-bank
Initializes Memory Bank with all core files.
Parameters:
rootPath
: Project root directory pathforce
(optional): Force re-initialization
Returns: Created files list and next steps guidance
get-memory-bank-info
Reads and returns all Memory Bank content (similar to codelf's get-project-info).
Parameters:
rootPath
: Project root directory path
Returns: Formatted Memory Bank content for AI context
update-memory-bank
Provides guidance for updating Memory Bank files.
Parameters:
rootPath
: Project root directory pathchangeType
: Type of change (architecture/feature/bugfix/refactor/decision/progress)description
: Brief description of the change
Returns: Detailed update instructions with templates and timestamps
Integration Tips
Cursor Setup
Add to Settings → Rules → User Rules:
Before starting any task, check if memory-bank directory exists in the project. If not, run the MCP command init-memory-bank.
Use the MCP command get-memory-bank-info to read Memory Bank content at session start.
After completing tasks or conversations, you must use the MCP command update-memory-bank to update Memory Bank content.
Follow the MCP guidance to maintain Memory Bank files.
Windsurf Setup
Add to Settings → Cascade → Memories and Rules → Global Rules:
Before starting any task, check if memory-bank directory exists in the project. If not, run the MCP command init-memory-bank.
Use the MCP command get-memory-bank-info to read Memory Bank content at session start.
After completing tasks or conversations, you must use the MCP command update-memory-bank to update Memory Bank content.
Follow the MCP guidance to maintain Memory Bank files.
Contributing
Contributions are welcome! Please feel free to submit issues or pull requests.
License
MIT
Acknowledgments
Inspired by the SPARC methodology and codelf.
Recommended Servers
playwright-mcp
A Model Context Protocol server that enables LLMs to interact with web pages through structured accessibility snapshots without requiring vision models or screenshots.
Magic Component Platform (MCP)
An AI-powered tool that generates modern UI components from natural language descriptions, integrating with popular IDEs to streamline UI development workflow.
Audiense Insights MCP Server
Enables interaction with Audiense Insights accounts via the Model Context Protocol, facilitating the extraction and analysis of marketing insights and audience data including demographics, behavior, and influencer engagement.

VeyraX MCP
Single MCP tool to connect all your favorite tools: Gmail, Calendar and 40 more.
graphlit-mcp-server
The Model Context Protocol (MCP) Server enables integration between MCP clients and the Graphlit service. Ingest anything from Slack to Gmail to podcast feeds, in addition to web crawling, into a Graphlit project - and then retrieve relevant contents from the MCP client.
Kagi MCP Server
An MCP server that integrates Kagi search capabilities with Claude AI, enabling Claude to perform real-time web searches when answering questions that require up-to-date information.

E2B
Using MCP to run code via e2b.
Neon Database
MCP server for interacting with Neon Management API and databases
Exa Search
A Model Context Protocol (MCP) server lets AI assistants like Claude use the Exa AI Search API for web searches. This setup allows AI models to get real-time web information in a safe and controlled way.
Qdrant Server
This repository is an example of how to create a MCP server for Qdrant, a vector search engine.