mcp-node-expert

saralegui-solutions/mcp-node-expert

3.2

If you are the rightful owner of mcp-node-expert and would like to certify it and/or have it hosted online, please leave a comment on the right or send an email to henry@mcphub.com.

A sophisticated Model Context Protocol (MCP) server providing comprehensive Node.js and NPM expertise for code validation, best practices analysis, and intelligent auto-fixing capabilities.

Tools
6
Resources
0
Prompts
0

Node.js Expert MCP Server

A sophisticated Model Context Protocol (MCP) server providing comprehensive Node.js and NPM expertise for code validation, best practices analysis, and intelligent auto-fixing capabilities. Features advanced learning integration with autonomous pattern recognition and cross-server knowledge sharing.

🌟 Features

🔍 Core Validation & Analysis

  • Project Validation: Complete Node.js project configuration analysis
  • Code Review: JavaScript/TypeScript code quality assessment
  • Module System Validation: CommonJS, ESM, and mixed module system analysis
  • TypeScript Configuration: Comprehensive tsconfig.json validation
  • Package.json Analysis: Dependency management and configuration validation
  • Best Practices: Context-aware recommendations and improvements

🧠 Learning Integration

  • Autonomous Learning: Records all validation and fix operations
  • Pattern Recognition: Learns from repeated issues and successful fixes
  • Knowledge Enhancement: Applies learned insights to improve validation results
  • Cross-server Communication: Integrates with MCP Self-Learning Server
  • Performance Optimization: Continuously improves based on usage patterns

🛠️ Auto-fixing Capabilities

  • Safe Fixes: Automatic application of low-risk corrections
  • Backup Creation: Automatic file backups before modifications
  • Smart Suggestions: Context-aware fix recommendations
  • Batch Operations: Multiple issue fixes in single operation

🔌 Integration Ready

  • Claudio Orchestration: Full integration with multi-agent coordination
  • MessageBus Communication: Real-time learning event broadcasting
  • Claude Desktop: Seamless Claude Code integration
  • Non-blocking Architecture: Graceful degradation when services unavailable

🚀 Quick Start

Prerequisites

  • Node.js 18.0.0 or higher
  • npm or yarn package manager
  • TypeScript 5.6.0 or higher

Installation

  1. Clone and Navigate

    cd /home/ben/saralegui-solutions-llc/mcp-servers/node-expert

  2. Install Dependencies

    npm install

  3. Build the Server

    npm run build

  4. Configure Claude Desktop Add to your Claude Desktop claude_desktop_config.json:

    {
  "mcpServers": {
    "node-expert": {
      "command": "node",
      "args": ["/home/ben/saralegui-solutions-llc/mcp-servers/node-expert/dist/index.js"]
    }
  }
}

  5. Test Installation

    npm test

🛠️ Available Tools

validate_project

Validates entire Node.js project configuration and structure.

Parameters:

  • projectPath (required): Path to the Node.js project directory
  • autoFix (optional): Automatically apply safe fixes (default: false)
  • includeOptional (optional): Include optional best practice suggestions (default: true)

Example:

{
  "projectPath": "/path/to/your/project",
  "autoFix": false,
  "includeOptional": true
}

review_code

Reviews JavaScript/TypeScript code for common issues and best practices.

Parameters:

  • code (required): Code content to review
  • fileName (optional): Name or path of the file
  • context (optional): Additional context (moduleSystem, hasTypeScript, projectPath)

Example:

{
  "code": "const fs = require('fs');\nimport path from 'path';",
  "fileName": "example.js",
  "context": {
    "moduleSystem": "mixed",
    "hasTypeScript": false
  }
}

suggest_fixes

Provides specific fixes for identified issues.

Parameters:

  • issues (required): Array of issues from validation
  • code (optional): Original code content
  • preferences (optional): Fix preferences (preferEsm, strictMode, modernSyntax)

auto_fix

Automatically applies safe fixes to code or configuration.

Parameters:

  • filePath (required): Path to file to fix
  • issues (required): Array of specific issues to fix
  • backup (optional): Create backup before fixing (default: true)

check_compatibility

Checks dependency compatibility and Node.js version requirements.

Parameters:

  • packageJsonPath (required): Path to package.json file
  • nodeVersion (optional): Target Node.js version (e.g., "18.0.0")

get_best_practices

Retrieves best practices and recommendations for Node.js development.

Parameters:

  • category (optional): Specific category (module-system, typescript, package-json, dependencies, all)
  • level (optional): Experience level (beginner, intermediate, advanced, all)

📊 Learning Features

Event Recording

Automatically records:

  • Tool usage patterns
  • Success/failure rates
  • Performance metrics
  • Context information
  • User preferences

Pattern Recognition

  • Identifies common issue patterns
  • Learns successful fix strategies
  • Adapts recommendations based on project context
  • Improves validation accuracy over time

Knowledge Sharing

  • Broadcasts learning events to MCP Self-Learning Server
  • Receives insights from other MCP servers
  • Updates local knowledge base with learned patterns
  • Enhances validation results with community knowledge

⚙️ Configuration

Environment Variables

  • LEARNING_ENABLED: Enable/disable learning features (default: true)
  • CLAUDIO_ENABLED: Enable/disable Claudio integration (default: true)
  • MESSAGE_BUS_URL: MessageBus WebSocket URL (default: ws://localhost:3001)
  • CLAUDIO_URL: Claudio orchestrator URL (default: http://localhost:3001)
  • LEARNING_BATCH_SIZE: Learning event batch size (default: 10)
  • LEARNING_FLUSH_INTERVAL: Event flush interval in ms (default: 30000)

Learning Configuration

# Enable full learning integration
LEARNING_ENABLED=true
CLAUDIO_ENABLED=true
MESSAGE_BUS_URL=ws://localhost:3001

# Disable learning (standalone mode)
LEARNING_ENABLED=false

🧪 Testing

Run All Tests

npm test

Test Learning Integration

./scripts/test-learning.sh

Test Server Integration

./scripts/test-server.sh

Manual Testing

# Test with Claude Desktop
# Use any of the 6 tools through Claude Desktop interface

# Direct MCP testing
echo '{"method":"tools/list","params":{},"id":1,"jsonrpc":"2.0"}' | node dist/index.js

🏗️ Architecture

Core Components

  • MCP Server: Main Model Context Protocol server implementation
  • Validators: Specialized validators for different aspects (module system, TypeScript, package.json)
  • Analyzers: Code analysis engines for pattern detection
  • Auto-fixers: Intelligent code fixing mechanisms
  • Knowledge Base: Local storage for best practices and patterns
  • Learning Integration: Autonomous learning and pattern recognition
  • Claudio Bridge: Communication with orchestration system

Learning Flow

  1. Event Capture: All tool interactions are recorded
  2. Pattern Analysis: Learning integration analyzes usage patterns
  3. Knowledge Update: Local knowledge base is updated with insights
  4. Result Enhancement: Validation results are improved with learned patterns
  5. Cross-server Sharing: Insights are shared with other MCP servers

🤝 Integration

With MCP Self-Learning Server

# Start the learning system
./scripts/start-learning-system.sh

With Claudio Orchestrator

The server automatically registers with Claudio when available and participates in:

  • Multi-agent workflows
  • Cross-server knowledge sharing
  • Performance optimization
  • Learning coordination

Standalone Usage

The server operates fully independently when learning services are unavailable:

  • All 6 tools function normally
  • Local knowledge base provides recommendations
  • Graceful degradation ensures reliability

📈 Performance

Metrics Tracked

  • Tool execution time
  • Validation accuracy
  • Fix success rate
  • Memory usage
  • Learning event processing

Optimization Features

  • Non-blocking learning initialization
  • Asynchronous event processing
  • Local caching of best practices
  • Efficient TypeScript compilation

🛡️ Security

Safe Operations

  • All auto-fixes create backups
  • No sensitive data in learning events
  • Path sanitization in event recording
  • Graceful error handling

Data Privacy

  • User paths are sanitized before sharing
  • No sensitive file content in logs
  • Local knowledge base encryption ready
  • Opt-out learning configuration

📝 Examples

See the examples/ directory for:

  • Basic project validation
  • Code review workflows
  • Auto-fixing examples
  • Learning integration demos
  • Claude Desktop usage patterns

🔧 Development

Building from Source

git clone <repository>
cd node-expert
npm install
npm run build

Development Mode

npm run dev

Linting

npm run lint

🤝 Contributing

See CONTRIBUTING.md for development guidelines and contribution process.

📞 Support

  • Documentation: See docs/ directory for detailed guides
  • Issues: Report issues through the issue tracker
  • Integration Help: See INTEGRATION.md for setup assistance

Saralegui Solutions LLC - Autonomous AI Development Tools