READMEmain

MCP Setup for Cursor IDE

This package provides automated setup for the Model Context Protocol (MCP) Memory Server in Cursor IDE. The MCP enables persistent memory across AI assistant conversations, project-specific rules, and enhanced code assistance.

Quick Start

Choose one of the following methods to run the setup script:

Windows

Double-click the setup-mcp.bat file or run it from the command prompt:

setup-mcp.bat

macOS / Linux

Run the setup script from the terminal:

chmod +x setup-mcp.sh
./setup-mcp.sh

Manual (Any Platform)

If you prefer to run the JavaScript script directly:

node setup-mcp.js

What This Script Does

Installs MCP Memory Server locally as a dev dependency to your project
Analyzes your project to determine its type, frameworks, and patterns
Extracts context from your project documentation
Configures Cursor IDE to use the MCP server
Creates project-specific rules based on your tech stack
Sets up a memory file within your project structure
Creates memory utilities for backup and exploration
Optimizes Cursor workspace settings for AI assistance

Configuration

The setup creates a .mcp directory in your project with:

ai_memory.json - The memory file that stores conversations
backups/ - Directory for memory backups
backup-memory.js - Utility script for creating backups
memory-explorer.js - Simple web viewer for memory entries

It also creates a .cursor/rules directory with:

project-rules.mdc - Main project rules and context
auth-rules.mdc - Authentication guidelines (if detected)
data-management-rules.mdc - Data management guidelines (if detected)
ui-components-rules.mdc - UI component guidelines (if detected)
best-practices.mdc - Best practices for your tech stack

After Setup

Restart Cursor IDE to apply the configuration
Start the MCP server with npm run mcp
Start coding with enhanced AI assistance and memory persistence
Create backups with npm run mcp:backup periodically
View memory entries with npm run mcp:explore

Memory File Structure

The memory file stores structured "memories" in JSON format:

{
  "entries": [
    {
      "id": "initial-setup",
      "timestamp": "2023-05-10T12:34:56.789Z",
      "content": "MCP Memory Server successfully configured for project-name."
    },
    {
      "id": "project-context",
      "timestamp": "2023-05-10T12:34:56.789Z",
      "content": "Project documentation and context extracted from README.md and other sources."
    },
    // Additional entries will be added as you work with the AI assistant
  ]
}

Project Rules

The rules files tell Claude how to understand and work with your codebase. They include:

Project structure information
Technical stack details
Coding patterns
Best practices
Memory utilization guidelines

Customizing Rules

You can customize any of the generated rule files in the .cursor/rules directory to better match your specific project requirements or preferences.

Troubleshooting

Node.js Required: Make sure Node.js is installed on your system
Permissions: Ensure you have write permissions in the project directory
Configuration Not Applied: Make sure to restart Cursor after setup
Memory Not Updating: Check if the MCP server is running with npm run mcp
Package.json Issues: If you encounter issues with package.json, run npm init -y before setup

Future Directions

The MCP setup process has several planned enhancements for future versions:

Team Collaboration

Shared team memories across multiple developers
Memory conflict resolution for collaborative editing
Memory versioning aligned with git branches
Team-specific rules and preferences

Continuous Learning

Periodic update of rules based on evolving codebase
Integration with CI/CD to update memories with each build
Automatic extraction of new project patterns
Learning from code review comments

IDE Integration

Deeper integration with Cursor IDE features
Memory visualization tools within the IDE
Rule editing interface with syntax highlighting
AI-assisted rule generation and refinement

Advanced Analysis

Code quality insights based on memory patterns
Performance optimization suggestions
Security vulnerability detection
Pattern-based refactoring suggestions

Multi-Project Context

Context sharing across related projects
Workspace-level memories for multi-project solutions
Cross-project pattern recognition
Organizational knowledge base integration

License

MIT

mcp-setup-tools