MCP Neurolora

View original on GitHub 

An intelligent MCP server that provides tools for code analysis using OpenAI API, code collection, and documentation generation.

🚀 Installation Guide

Don’t worry if you don’t have anything installed yet! Just follow these steps or ask your assistant to help you with the installation.

Step 1: Install Node.js

macOS

1. Install Homebrew if not installed:

   /bin/bash -c "$(curl -fsSL https://raw.githubusercontent.com/Homebrew/install/HEAD/install.sh)"

2. Install Node.js 18:

   brew install node@18
   echo 'export PATH="/opt/homebrew/opt/node@18/bin:$PATH"' >> ~/.zshrc
   source ~/.zshrc

Windows

1. Download Node.js 18 LTS from nodejs.org 
2. Run the installer
3. Open a new terminal to apply changes

Linux (Ubuntu/Debian)

curl -fsSL https://deb.nodesource.com/setup_18.x | sudo -E bash -
sudo apt-get install -y nodejs

Step 2: Install uv and uvx

All Operating Systems

1. Install uv:

   curl -LsSf https://astral.sh/uv/install.sh | sh

2. Install uvx:

   uv pip install uvx

Step 3: Verify Installation

Run these commands to verify everything is installed:

node --version  # Should show v18.x.x
npm --version   # Should show 9.x.x or higher
uv --version    # Should show uv installed
uvx --version   # Should show uvx installed

Step 4: Configure MCP Server

Your assistant will help you:

1. Find your Cline settings file:

   • VSCode: ~/Library/Application Support/Code/User/globalStorage/saoudrizwan.claude-dev/settings/cline_mcp_settings.json
   • Claude Desktop: ~/Library/Application Support/Claude/claude_desktop_config.json
   • Windows VSCode: %APPDATA%/Code/User/globalStorage/saoudrizwan.claude-dev/settings/cline_mcp_settings.json
   • Windows Claude: %APPDATA%/Claude/claude_desktop_config.json

2. Add this configuration:

   {
     "mcpServers": {
       "aindreyway-mcp-neurolora": {
         "command": "npx",
         "args": ["-y", "@aindreyway/mcp-neurolora@latest"],
         "env": {
           "NODE_OPTIONS": "--max-old-space-size=256",
           "OPENAI_API_KEY": "your_api_key_here"
         }
       }
     }
   }

Step 5: Install Base Servers

Simply ask your assistant: “Please install the base MCP servers for my environment”

Your assistant will:

1. Find your settings file
2. Run the install_base_servers tool
3. Configure all necessary servers automatically

After the installation is complete:

1. Close VSCode completely (Cmd+Q on macOS, Alt+F4 on Windows)
2. Reopen VSCode
3. The new servers will be ready to use

Important: A complete restart of VSCode is required after installing the base servers for them to be properly initialized.

Note: This server uses npx for direct npm package execution, which is optimal for Node.js/TypeScript MCP servers, providing seamless integration with the npm ecosystem and TypeScript tooling.

Base MCP Servers

The following base servers will be automatically installed and configured:

• fetch: Basic HTTP request functionality for accessing web resources
• puppeteer: Browser automation capabilities for web interaction and testing
• sequential-thinking: Advanced problem-solving tools for complex tasks
• github: GitHub integration features for repository management
• git: Git operations support for version control
• shell: Basic shell command execution with common commands:
  • ls: List directory contents
  • cat: Display file contents
  • pwd: Print working directory
  • grep: Search text patterns
  • wc: Count words, lines, characters
  • touch: Create empty files
  • find: Search for files

🎯 What Your Assistant Can Do

Ask your assistant to:

• “Analyze my code and suggest improvements”
• “Install base MCP servers for my environment”
• “Collect code from my project directory”
• “Create documentation for my codebase”
• “Generate a markdown file with all my code”

🛠 Available Tools

analyze_code

Analyzes code using OpenAI API and generates detailed feedback with improvement suggestions.

Parameters:

• codePath (required): Path to the code file or directory to analyze

Example usage:

{
  "codePath": "/path/to/your/code.ts"
}

The tool will:

1. Analyze your code using OpenAI API
2. Generate detailed feedback with:
   • Issues and recommendations
   • Best practices violations
   • Impact analysis
   • Steps to fix
3. Create two output files in your project:
   • LAST_RESPONSE_OPENAI.txt - Human-readable analysis
   • LAST_RESPONSE_OPENAI_GITHUB_FORMAT.json - Structured data for GitHub issues

Note: Requires OpenAI API key in environment configuration

collect_code

Collects all code from a directory into a single markdown file with syntax highlighting and navigation.

Parameters:

• directory (required): Directory path to collect code from
• outputPath (optional): Path where to save the output markdown file
• ignorePatterns (optional): Array of patterns to ignore (similar to .gitignore)

Example usage:

{
  "directory": "/path/to/project/src",
  "outputPath": "/path/to/project/src/FULL_CODE_SRC_2024-12-20.md",
  "ignorePatterns": ["*.log", "temp/", "__pycache__", "*.pyc", ".git"]
}

install_base_servers

Installs base MCP servers to your configuration file.

Parameters:

• configPath (required): Path to the MCP settings configuration file

Example usage:

{
  "configPath": "/path/to/cline_mcp_settings.json"
}

🔧 Features

The server provides:

• Code Analysis:

  • OpenAI API integration
  • Structured feedback
  • Best practices recommendations
  • GitHub issues generation

• Code Collection:

  • Directory traversal
  • Syntax highlighting
  • Navigation generation
  • Pattern-based filtering

• Base Server Management:

  • Automatic installation
  • Configuration handling
  • Version management

📄 License

MIT License - feel free to use this in your projects!

👤 Author

Aindreyway

• GitHub: @aindreyway 

⭐️ Support

Give a ⭐️ if this project helped you!