Memory Bank MCP Server
Memory Bank MCP Server
A Model Context Protocol (MCP) server implementing the Memory Bank system based on the cursor-memory-bank custom modes.
Overview
This MCP server provides structured workflow tools for managing software development projects through different complexity levels and phases:
VAN Mode: Project initialization and complexity assessment
PLAN Mode: Detailed implementation planning
CREATIVE Mode: Design and architecture decisions
IMPLEMENT Mode: Code implementation and execution
REFLECT+ARCHIVE Mode: Project reflection and documentation archiving
Features
🎯 VAN Mode (Entry Point)
Initialize new projects with complexity levels 1-4
Set up Memory Bank file structure
Determine appropriate next mode based on complexity
📋 PLAN Mode (Planning)
Generate implementation plans based on complexity level
Create detailed step-by-step approaches
Identify components requiring creative phases
🎨 CREATIVE Mode (Design)
Architecture design with multiple options analysis
Algorithm design with complexity considerations
UI/UX design with accessibility and usability focus
Structured documentation of design decisions
⚒️ IMPLEMENT Mode (Execution)
Phase-based implementation for complex projects
Integration testing and documentation
Progress tracking and status updates
🤔 REFLECT+ARCHIVE Mode (Completion)
Implementation reflection and lessons learned
Complete project archiving
Preparation for next task initialization
Installation
npm install
npm run buildUsage
Running the Server
npm startAvailable Tools
1. van_mode
Initialize a project with complexity assessment.
Parameters:
complexity(required): "1" | "2" | "3" | "4"Level 1: Quick bug fix
Level 2: Simple enhancement
Level 3: Complex feature
Level 4: Major feature/refactor
task_description(optional): Description of the task
Example:
{
"complexity": "3",
"task_description": "Implement user authentication system"
}2. plan_mode
Create detailed implementation plan.
Parameters:
complexity(optional): Complexity level (reads from tasks.md if not provided)
3. creative_mode
Perform design and architecture work.
Parameters:
component_name(required): Name of component requiring designcreative_type(required): "architecture" | "algorithm" | "uiux"requirements(required): Requirements and constraints
Example:
{
"component_name": "user_auth_system",
"creative_type": "architecture",
"requirements": "Secure authentication with JWT tokens, OAuth integration, and role-based access control"
}4. implement_mode
Execute implementation based on plan.
Parameters:
phase(optional): Implementation phase for complex projectscomplexity(optional): Complexity level (reads from tasks.md if not provided)
5. reflect_archive_mode
Reflect on implementation and archive documentation.
Parameters:
action(required): "reflect" | "archive""reflect": Review and document implementation experience
"archive": Create final documentation archive
Memory Bank Files
The server creates and manages several files in the .memory_bank directory:
tasks.md: Main task tracking and statusactiveContext.md: Current context and focusprogress.md: Implementation progress trackingimplementation-plan.md: Detailed implementation planreflection.md: Post-implementation reflectioncreative-{component}.md: Creative phase documentationdocs/archive/project-archive.md: Final project archive
Workflow Example
# 1. Initialize project
van_mode(complexity="3", task_description="Build REST API")
# 2. Create implementation plan
plan_mode()
# 3. Design complex components (if Level 3-4)
creative_mode(component_name="api_architecture", creative_type="architecture", requirements="...")
# 4. Implement the solution
implement_mode(phase="core_components")
# 5. Reflect and archive
reflect_archive_mode(action="reflect")
reflect_archive_mode(action="archive")Complexity Levels
Level 1: Quick bug fixes - Direct implementation, minimal planning
Level 2: Simple enhancements - Streamlined planning, straightforward implementation
Level 3: Complex features - Comprehensive planning, creative phases may be required
Level 4: Major features/refactors - Detailed architecture, phased implementation, creative design required
Development
# Development mode with auto-reload
npm run dev
# Build TypeScript
npm run build
# Run compiled server
npm startTroubleshooting
Module Resolution Errors
If you encounter ERR_MODULE_NOT_FOUND errors when running the server, ensure:
Build First: Always run
npm run buildbeforenpm startES Module Imports: All imports in source files use
.jsextensionsTypeScript Config: The
tsconfig.jsonis configured for ES modules with"module": "ESNext"
# Clean build if issues persist
rm -rf dist/
npm run build
npm startMemory Bank Directory Permissions
If the server can't create the .memory_bank directory:
# Ensure write permissions in the working directory
chmod 755 .Integration
This MCP server can be integrated with:
Claude Desktop: Add to MCP server configuration
Cursor IDE: Use with MCP client capabilities
VS Code: Through MCP extensions
Custom clients: Any MCP-compatible application
Based On
This implementation is based on the excellent cursor-memory-bank project by vanzan01, adapted for use as an MCP server.
License
MIT
Latest Blog Posts
- Who's Calling? MCP Hosts Are an Identity Blind Spot (And the Spec Knows It)By Om-Shree-0709 on .mcpAgent IdentityOAuth 2.1
- Your AI Chatbot Just Exposed Your CEO's Salary to an InternBy Om-Shree-0709 on .Agent IdentityMCP SecurityOAuth Delegation
- Why MCP Servers Need Execution Sandboxing (And Why Your Current Stack Isn't Enough)By Om-Shree-0709 on .Agentic AiPrompt InjectionWebAssembly
MCP directory API
We provide all the information about MCP servers via our MCP API.
curl -X GET 'https://glama.ai/api/mcp/v1/servers/nicavcrm/memory-bank-mcp'
If you have feedback or need assistance with the MCP directory API, please join our Discord server