MCP Personal Assistant

# Remote Database Setup Guide This guide shows how to configure the MCP Personal Assistant to use a remote database instead of a local SQLite file. ## Why Use a Remote Database? - **Multi-device sync**: Access your data from multiple computers - **Backup & reliability**: Professional database hosting with automated backups - **Scalability**: Better performance for large datasets - **Collaboration**: Share data across team members (future feature) ## Option 1: PostgreSQL (Recommended) ### Supported Services - **Supabase** (Free tier available, easiest setup) - **AWS RDS** (Enterprise-grade) - **Google Cloud SQL** - **Azure Database for PostgreSQL** - **PlanetScale** (MySQL-compatible) - **Neon** (Serverless PostgreSQL) ### Setup Instructions #### 1. Install PostgreSQL Dependencies ```bash cd /path/to/your/mcp-pa source venv/bin/activate pip install asyncpg psycopg2-binary ``` #### 2. Create Database (Supabase Example) 1. Go to [supabase.com](https://supabase.com) and create an account 2. Create a new project 3. Go to Settings > Database 4. Copy your connection string (it looks like): ``` postgresql://postgres:[YOUR-PASSWORD]@db.[YOUR-PROJECT].supabase.co:5432/postgres ``` #### 3. Update Claude Desktop Configuration Replace your `claude_desktop_config.json` with: ```json { "mcpServers": { "personal-assistant": { "command": "/path/to/your/mcp-pa/venv/bin/python", "args": ["-m", "src.server"], "cwd": "/path/to/your/mcp-pa", "env": { "MCP_PA_DB_TYPE": "postgresql", "MCP_PA_PGVECTOR_CONNECTION_STRING": "postgresql://postgres:YOUR-PASSWORD@db.YOUR-PROJECT.supabase.co:5432/postgres" } } } } ``` #### 4. Test the Connection ```bash source venv/bin/activate python examples/basic_usage.py ``` If successful, your data is now stored in the remote PostgreSQL database! ### Advanced PostgreSQL Features The PostgreSQL backend includes: - **Vector embeddings**: For semantic search (future feature) - **Full-text search**: Across documents and projects - **JSONB fields**: For flexible metadata storage - **Concurrent access**: Multiple users/devices simultaneously - **Transactions**: Data integrity guarantees ## Option 2: Cloud-Synced SQLite (Simple) If you prefer to keep using SQLite but want cross-device sync: ### Using Cloud Storage ```json { "mcpServers": { "personal-assistant": { "command": "/path/to/your/mcp-pa/venv/bin/python", "args": ["-m", "src.server"], "cwd": "/path/to/your/mcp-pa", "env": { "MCP_PA_DB_TYPE": "sqlite", "MCP_PA_DB_PATH": "/Users/your-name/Dropbox/mcp-pa/database.sqlite" } } } } ``` **Supported cloud storage:** - Dropbox: `/Users/your-name/Dropbox/mcp-pa/` - iCloud: `/Users/your-name/Library/Mobile Documents/com~apple~CloudDocs/mcp-pa/` - Google Drive: `/Users/your-name/Google Drive/mcp-pa/` ⚠️ **Note**: SQLite with cloud sync can have conflicts if accessed simultaneously from multiple devices. ## Connection String Examples ### PostgreSQL Formats ```bash # Supabase postgresql://postgres:password@db.project.supabase.co:5432/postgres # AWS RDS postgresql://username:password@mydb.cluster-xxx.us-east-1.rds.amazonaws.com:5432/mcppa # Google Cloud SQL postgresql://username:password@127.0.0.1:5432/mcppa?host=/cloudsql/project:region:instance # Local PostgreSQL postgresql://postgres:password@localhost:5432/mcp_personal_assistant # With SSL (recommended for production) postgresql://user:pass@host:5432/db?sslmode=require ``` ## Environment Variables You can also set these as environment variables instead of in the config: ```bash export MCP_PA_DB_TYPE=postgresql export MCP_PA_PGVECTOR_CONNECTION_STRING="postgresql://postgres:password@host:5432/db" # Then use in Claude Desktop config: { "mcpServers": { "personal-assistant": { "command": "/path/to/your/mcp-pa/venv/bin/python", "args": ["-m", "src.server"], "cwd": "/path/to/your/mcp-pa" } } } ``` ## Migration from SQLite To migrate your existing SQLite data to PostgreSQL: 1. **Export your SQLite data:** ```bash source venv/bin/activate python scripts/export_data.py --format json --output backup.json ``` 2. **Switch to PostgreSQL** (follow setup above) 3. **Import your data:** ```bash python scripts/import_data.py --format json --input backup.json ``` *Note: Migration scripts need to be created if not already available.* ## Troubleshooting ### Connection Issues 1. **Check firewall settings** - ensure port 5432 is accessible 2. **Verify connection string** - test with psql command-line tool 3. **Check SSL requirements** - some hosts require `sslmode=require` ### Performance Tips 1. **Connection pooling** is automatically handled 2. **Indexes** are created automatically for common queries 3. **Consider read replicas** for high-traffic scenarios ### Security Best Practices 1. **Use environment variables** for credentials (not config files) 2. **Enable SSL/TLS** for production databases 3. **Restrict IP access** in your database hosting settings 4. **Use strong passwords** and rotate them regularly ## Cost Considerations - **Supabase**: Free tier (500MB), then ~$25/month - **AWS RDS**: ~$15-30/month for small instances - **Google Cloud SQL**: ~$10-25/month for basic setup - **Neon**: Generous free tier, serverless pricing The SQLite + cloud storage option costs just the price of your cloud storage plan.