@corbyjames/persistent-context-mcp

AI memory management and context persistence system with Model Context Protocol (MCP) server for Claude Desktop and other AI assistants.

🌟 Features

• Persistent Context Storage: Store and retrieve conversation contexts using Neo4j graph database
• MCP Server Integration: Full compatibility with Claude Desktop via Model Context Protocol
• Web UI: Interactive dashboard for managing contexts and visualizing relationships
• Multi-Storage Support: Neo4j for graph relationships, file system for backups
• AI Integration: Optional OpenAI/Anthropic API support for semantic search
• Docker/Podman Support: Containerized deployment options
• CLI Tool: Global command-line interface for easy management

🚀 Quick Start

Installation

# Install globally
npm install -g @corbyjames/persistent-context-mcp

# Or using npx (no installation required)
npx @corbyjames/persistent-context-mcp init

Initialize Configuration

# Interactive setup
persistent-context init

# With Claude Desktop integration
persistent-context init --claude

# With Docker configuration
persistent-context init --docker

Start Services

# Start Neo4j database
persistent-context neo4j start

# Start MCP server
persistent-context start

# Start with UI
persistent-context start --ui

# Or run as daemon
persistent-context start --daemon

🔧 Claude Desktop Integration

Automatic Setup

persistent-context claude-setup

This will automatically configure Claude Desktop to use the MCP server.

Manual Configuration

Add to ~/Library/Application Support/Claude/claude_desktop_config.json:

{
  "mcpServers": {
    "persistent-context": {
      "command": "persistent-context",
      "args": ["start"],
      "env": {
        "PORT": "3000"
      }
    }
  }
}

Note: Restart Claude Desktop after configuration.

🐳 Docker/Podman Deployment

Using Docker

# Generate docker-compose.yml
persistent-context container docker

# Start services
docker-compose up -d

# View logs
docker-compose logs -f

Using Podman

# Generate podman-compose.yml
persistent-context container podman

# Start services
podman-compose up -d

# View logs
podman-compose logs -f

Pre-built Docker Image

docker run -d \
  --name persistent-context \
  -p 3000:3000 \
  -p 5173:5173 \
  -e NEO4J_URI=bolt://neo4j:7687 \
  -e NEO4J_USER=neo4j \
  -e NEO4J_PASSWORD=password \
  corbyjames/persistent-context-mcp:latest

📁 Configuration

Environment Variables

Create a .env file in your project directory:

# Neo4j Configuration
NEO4J_URI=bolt://localhost:7687
NEO4J_USER=neo4j
NEO4J_PASSWORD=password

# Server Configuration
PORT=3000
NODE_ENV=production
LOG_LEVEL=info

# Data Storage
DATA_DIR=~/.persistent-context
BACKUP_DIR=~/.persistent-context/backups

# Optional: AI Integration
OPENAI_API_KEY=your-api-key

# Security - Store in secure vault for production
SECRETS_PATH=~/.persistent-context/secrets

Secrets Management

For production deployments, we recommend storing secrets using:

• AWS Secrets Manager

  aws secretsmanager create-secret \
    --name persistent-context/neo4j \
    --secret-string '{"password":"secure-password"}'
  

• Azure Key Vault

  az keyvault secret set \
    --vault-name mykeyvault \
    --name neo4j-password \
    --value "secure-password"
  

• HashiCorp Vault

  vault kv put secret/persistent-context \
    neo4j_password="secure-password"
  

• Kubernetes Secrets

  kubectl create secret generic persistent-context \
    --from-literal=neo4j-password=secure-password
  

🖥️ CLI Commands

Core Commands

# Initialize configuration
persistent-context init [options]
  --claude    Setup for Claude Desktop
  --docker    Generate Docker config
  --podman    Generate Podman config

# Start MCP server
persistent-context start [options]
  -d, --daemon    Run as daemon
  --ui            Also start the UI

# Start web UI
persistent-context ui [options]
  -p, --port    UI port (default: 5173)

# Manage Neo4j
persistent-context neo4j <action>
  start    Start Neo4j container
  stop     Stop Neo4j container
  status   Check Neo4j status

# Setup Claude Desktop
persistent-context claude-setup

# Generate container config
persistent-context container <runtime>
  docker    Generate docker-compose.yml
  podman    Generate podman-compose.yml

Short Alias

You can also use the short alias pcmcp:

pcmcp init
pcmcp start
pcmcp ui

🏗️ Architecture

┌─────────────────┐     ┌──────────────┐     ┌─────────────┐
│  Claude Desktop │────▶│  MCP Server  │────▶│    Neo4j    │
└─────────────────┘     └──────────────┘     └─────────────┘
                               │
                               ▼
                        ┌──────────────┐
                        │   Web UI     │
                        └──────────────┘

Components

• MCP Server: Implements Model Context Protocol for AI assistants
• Neo4j Database: Stores contexts as a knowledge graph
• Web UI: React-based dashboard for visualization
• CLI Tool: Command-line interface for management

🔌 MCP Tools Available

When integrated with Claude Desktop, the following tools are available:

save_context

Save a new context to the persistent store.

{
  title: string;
  content: string;
  tags?: string[];
  metadata?: Record<string, any>;
}

search_contexts

Search for contexts using semantic or keyword search.

{
  query: string;
  limit?: number;
  tags?: string[];
  dateRange?: {
    start: string;
    end: string;
  };
}

generate_template

Generate a context template for specific use cases.

{
  type: 'technical' | 'business' | 'analysis' | 'planning';
  domain?: string;
  includeExamples?: boolean;
}

📊 Web UI Features

Access the web UI at http://localhost:5173 (when running with --ui flag):

• Dashboard: Overview of stored contexts and statistics
• Context Browser: Search and browse all contexts
• Graph Visualization: Interactive knowledge graph
• Templates: Pre-built context templates
• Analytics: Usage patterns and insights
• Settings: Configuration management

🔒 Security Considerations

1. Default Credentials: Change default Neo4j password immediately
2. Network Security: Use TLS/SSL in production
3. Access Control: Implement authentication for web UI
4. Data Encryption: Enable encryption at rest for Neo4j
5. Secrets Management: Use proper secret vaults in production

🧪 Testing

# Run all tests
npm test

# Unit tests only
npm run test:unit

# Integration tests
npm run test:integration

# E2E tests
npm run test:e2e

# Test coverage
npm run test:coverage

📦 Building from Source

# Clone repository
git clone https://github.com/corbyjames/persistent-context-mcp.git
cd persistent-context-mcp

# Install dependencies
npm install

# Build project
npm run build

# Run locally
npm start

🤝 Contributing

Contributions are welcome! Please read our Contributing Guide for details.

1. Fork the repository
2. Create your feature branch (git checkout -b feature/amazing-feature)
3. Commit your changes (git commit -m 'Add amazing feature')
4. Push to the branch (git push origin feature/amazing-feature)
5. Open a Pull Request

📄 License

This project is licensed under the MIT License - see the LICENSE file for details.

🙏 Acknowledgments

• Anthropic for Claude and MCP protocol
• Neo4j for the graph database
• IBM TargetProcess for project management inspiration
• Open source community for various dependencies

📞 Support

• Issues: GitHub Issues
• Discussions: GitHub Discussions
• Email: corby.james@example.com

🗺️ Roadmap

• [ ] Multi-user support with authentication
• [ ] Cloud deployment templates (AWS, Azure, GCP)
• [ ] Additional AI model integrations
• [ ] Export/Import functionality
• [ ] Context versioning and history
• [ ] Collaborative context editing
• [ ] Mobile app support

---

Made with ❤️ by Corby James