Neural_MCP

Toowiredd/Neural_MCP

3.2

If you are the rightful owner of Neural_MCP 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.

Neural MCP is a Model Context Protocol server designed to provide intelligent conversation memory with features like automatic pattern detection and entity extraction.

Tools

  1. capture_conversation

    Intelligent conversation capture with automatic analysis.

  2. search_memory

    Smart search with relevance scoring and context.

  3. get_smart_context

    Contextual suggestions and related information.

  4. get_memory_status

    System intelligence statistics and health.

  5. get_repositories

    Auto-created knowledge repositories.

Neural MCP - Intelligent Conversation Memory

Generated by Copilot

🧠 External Memory That Works Like Your Own Thoughts

Neural MCP is a production-ready Model Context Protocol (MCP) server that provides intelligent conversation memory with automatic pattern detection, entity extraction, and neurodivergent-optimized workflow support.

Built for developers who need external memory that works as naturally as their own thoughts.

✨ Core Features

🤖 Intelligent Memory Capture

  • Automatic Entity Extraction: Detects people, organizations, technologies, and projects
  • Smart Pattern Recognition: Identifies project discussions, brainstorming sessions, and workflows
  • Knowledge Graph Construction: Automatically builds interconnected knowledge structures
  • Context Preservation: Never loses conversation context or decision rationale

🧩 Optimized MCP Tools

  • capture_conversation - Intelligent conversation capture with automatic analysis
  • search_memory - Smart search with relevance scoring and context
  • get_smart_context - Contextual suggestions and related information
  • get_memory_status - System intelligence statistics and health
  • get_repositories - Auto-created knowledge repositories
  • get_conversation_context - Detailed context with entity relationships

🧠 Neurodivergent-Optimized Design

  • Zero Manual Organization: All categorization happens automatically
  • Executive Function Support: Reduces cognitive load through intelligent automation
  • ADHD-Friendly: Captures hyperfocus sessions and maintains momentum
  • Autism-Friendly: Consistent patterns and detailed context preservation
  • Decision Reduction: Automates organization and linking decisions

🚀 Quick Start

Installation

# Clone the repository
git clone https://github.com/toowiredd/Neural_MCP.git
cd Neural_MCP

# Install dependencies
npm install

# Start the Neural MCP server
npm start

MCP Client Integration

Add Neural MCP to your MCP client configuration:

{
  "mcpServers": {
    "neural-mcp": {
      "command": "node",
      "args": ["path/to/neural-mcp-server.js"]
    }
  }
}

Test the System

npm test

🔧 Configuration

Environment Variables

# Disable automatic intelligence (default: enabled)
export AUTO_INTELLIGENCE=false

# Change repository creation threshold (default: 3)
export REPO_THRESHOLD=5

# Custom memory storage location
export NEURAL_MEMORY_PATH="./custom-memory.jsonl"

Default Behavior

  • Automatic Intelligence: ✅ Enabled
  • Repository Threshold: 3 topic mentions
  • Storage Format: JSONL (git-friendly, append-only)
  • Entity Confidence: 0.7 minimum threshold
  • Context Results: 10 maximum per query

📊 How It Works

Automatic Intelligence Pipeline

Conversation Input
    ↓
Entity Extraction (NLP)
    ↓
Topic Detection
    ↓
Knowledge Graph Update
    ↓
Repository Auto-Creation
    ↓
Context Index Rebuild
    ↓
Smart Storage

Example: Project Detection

// Input conversation:
"I'm building a React app called TaskMaster using TypeScript and Node.js"

// Automatically detected:
{
  entities: [
    { name: "React", type: "technology" },
    { name: "TaskMaster", type: "project" },
    { name: "TypeScript", type: "technology" },
    { name: "Node.js", type: "technology" }
  ],
  topics: ["development", "frontend", "backend"],
  repositories: ["repo_react", "repo_typescript"] // auto-created if threshold reached
}

Smart Context Example

// Query: "CSS Grid layout"
// Neural MCP returns:
{
  relevantEntities: [
    { name: "React", type: "technology" },
    { name: "frontend_development", type: "topic" }
  ],
  relatedTopics: ["web design", "responsive design"],
  suggestedRepositories: ["repo_css", "repo_frontend"],
  contextualHints: ["This seems project-related. Consider breaking into smaller tasks."]
}

🛠️ Architecture

Technology Stack

  • Node.js with ES modules
  • @modelcontextprotocol/sdk for MCP compliance
  • natural for NLP and tokenization
  • compromise for advanced text analysis

Data Storage

  • JSONL format for efficient append-only operations
  • Knowledge graph structure with entities, relations, conversations
  • File-based storage scales to millions of conversations
  • Git-friendly text format for version control

Intelligence Processing

  • Real-time NLP analysis during conversation capture
  • Confidence scoring for all detected entities and relationships
  • Background pattern detection for repository auto-creation
  • Smart indexing for fast search and context retrieval

🧠 Neurodivergent Workflow Benefits

For ADHD

  • Hyperfocus Capture: Automatically organizes ideas during productive sessions
  • Interest Tracking: Maps evolving interests and project engagement
  • Momentum Maintenance: Visual progress through auto-created repositories
  • Context Switching: Smart suggestions help resume interrupted work

For Autism

  • Predictable Organization: Consistent naming and categorization patterns
  • Detailed Context: Complete conversation history and decision rationale
  • Special Interest Integration: Automatically links projects to areas of expertise
  • Reduced Overwhelm: No manual organization decisions required

For Executive Function Challenges

  • External Working Memory: Offloads cognitive burden to the system
  • Pattern Recognition: Identifies project vs. brainstorming vs. technical content
  • Automatic Categorization: No need to manually organize conversations
  • Context Restoration: Smart suggestions for resuming work

📈 Performance & Scalability

Optimizations

  • Efficient NLP processing with configurable confidence thresholds
  • Smart caching of topic frequencies and entity relationships
  • Minimal memory footprint through streaming operations
  • Fast search via indexed knowledge structures

Scalability Features

  • Append-only storage for optimal performance
  • Configurable intelligence can be tuned for different use cases
  • Modular design allows feature customization
  • Standard MCP protocol for universal AI client compatibility

🤝 Contributing

Neural MCP is designed for collaborative development:

Development Setup

  1. Fork the repository
  2. Create a feature branch
  3. Make your changes
  4. Add tests for new functionality
  5. Submit a pull request

Code Standards

  • ES modules with Node.js 18+
  • Comprehensive JSDoc documentation
  • Error handling for all async operations
  • MCP protocol compliance

📝 License

MIT License - Open source and freely usable

🆘 Support & Documentation

Troubleshooting

  • Ensure Node.js >=18.0.0
  • Check MCP client compatibility
  • Verify file system permissions for memory storage

Advanced Usage

  • See code comments for API details
  • Check MCP tool schemas for integration specifics
  • Review environment variables for customization options

Performance Tuning

  • Adjust REPO_THRESHOLD for your workflow
  • Configure AUTO_INTELLIGENCE based on needs
  • Use time-range filters for large memory stores

🎯 Mission

Neural MCP transforms AI conversations into persistent, intelligent knowledge that grows with you.

Perfect for neurodivergent developers, researchers, and anyone who needs external memory that works as naturally as their own thoughts. Zero configuration, maximum intelligence.

Built with ❤️ for minds that think differently.