🎥 MemVid MCP

🧠 AI Memory Bank Management with Enhanced Search

Transform your files into searchable AI memory banks using MP4 videos and vector embeddings. Seamlessly integrates with Cursor and Claude Desktop via the Model Context Protocol (MCP).

Features

• 🎥 MP4-based Memory Banks: Store and search content using video embeddings
• 🔍 Enhanced Search: Semantic search with filtering and sorting
• 🚀 Easy Setup: One-command installation via npx
• 🔧 Auto-Configuration: Automatically configures Cursor MCP settings
• 📁 Multiple Sources: Files, directories, URLs, and text content
• ⚡ High Performance: Sub-second search responses with caching

Quick Start

1. Install and Setup

npx @kcpatt27/memvid-mcp

This single command will:

• Install the MemVid MCP Server
• Check system requirements
• Configure Cursor automatically
• Set up Python dependencies

2. Start Using

1. Restart Cursor (if currently running)
2. Open any project in Cursor
3. Look for "memvid" in the MCP Tools section
4. Create your first memory bank using the create_memory_bank tool

Prerequisites

• Node.js 18+ - Download here
• Python 3.8+ - Download here
• Cursor - Download here

The setup command will help you install missing dependencies.

MCP Client Configuration

For Cursor, Claude Desktop, or other MCP clients:

{
  "mcpServers": {
    "memvid": {
      "command": "npx",
      "args": ["-y", "@kcpatt27/memvid-mcp", "--server"]
    }
  }
}

Custom Configuration

{
  "mcpServers": {
    "memvid": {
      "command": "npx",
      "args": ["-y", "@kcpatt27/memvid-mcp", "--server"],
      "env": {
        "MEMORY_BANKS_DIR": "/custom/path/to/banks",
        "PYTHON_EXECUTABLE": "python3"
      }
    }
  }
}

Available Tools

🏦 create_memory_bank

Create memory banks from various sources:

• Files: Text, code, documents
• Directories: Entire project folders
• URLs: Web content
• Text: Direct text input

🔍 search_memory

Advanced search with:

• Semantic matching
• File type filtering
• Content length filters
• Date range filtering
• Custom sorting

📋 list_memory_banks

List all available memory banks with metadata

➕ add_to_memory

Add new content to existing memory banks

🎯 get_context

Get formatted context for AI conversations

🏥 health_check & system_diagnostics

Monitor system health and performance

Commands

# Setup (default)
npx @kcpatt27/memvid-mcp

# Check system status
npx @kcpatt27/memvid-mcp --check

# Auto-install dependencies
npx @kcpatt27/memvid-mcp --install

# Show configuration
npx @kcpatt27/memvid-mcp --config

# Show help
npx @kcpatt27/memvid-mcp --help

# Show version
npx @kcpatt27/memvid-mcp --version

Usage Examples

Create a Memory Bank

Use the create_memory_bank tool in Cursor:

{
  "name": "my-project",
  "description": "My project documentation and code",
  "sources": [
    {
      "type": "directory",
      "path": "./src",
      "options": {
        "file_types": ["ts", "js", "md"]
      }
    },
    {
      "type": "file", 
      "path": "./README.md"
    }
  ],
  "tags": ["project", "documentation"]
}

Search Memory Banks

Use the search_memory tool:

{
  "query": "authentication and security",
  "memory_banks": ["my-project"],
  "top_k": 10,
  "filters": {
    "file_types": ["ts", "js"],
    "content_length": {
      "min": 100
    }
  },
  "sort_by": "relevance"
}

Troubleshooting

Setup Issues

# Check what's wrong
npx @kcpatt27/memvid-mcp --check

# Try auto-fix
npx @kcpatt27/memvid-mcp --install

Common Issues

1. Python not found: Install Python 3.8+ from python.org
2. MemVid not installed: Run pip install memvid
3. Cursor not detected: Ensure Cursor is installed and running
4. Permission errors: Run with appropriate permissions

Manual Python Setup

# Install MemVid package
pip install memvid

# Or with pip3
pip3 install memvid

Development

# Clone the repository
git clone https://github.com/kcpatt27/memvid-mcp.git
cd memvid-mcp

# Install dependencies
npm install

# Build
npm run build

# Test locally
node dist/cli.js --check

Environment Variables

• MEMORY_BANKS_DIR: Custom memory banks directory
• PYTHON_EXECUTABLE: Custom Python executable path
• MEMVID_CONFIG_PATH: Custom configuration file path
• LOG_LEVEL: Logging level (info, warn, error, debug)

Performance

• Memory Bank Creation: ~3-5 seconds
• Search Response: <500ms (cached)
• Memory Usage: <200MB baseline
• Concurrent Users: 5+ supported

License

MIT License - see LICENSE for details.

Support

• 📖 Documentation: GitHub Wiki
• 🐛 Issues: GitHub Issues
• 💬 Discussions: GitHub Discussions

Contributing

We welcome contributions! Please see CONTRIBUTING.md for guidelines.

Made with ❤️ for the AI community