WaniKani MCP Server

# Deployment Guide ## Render Deployment ### Prerequisites 1. GitHub repository with the code 2. Render account 3. PostgreSQL database ### Steps 1. **Create PostgreSQL Database on Render** - Go to Render Dashboard - Create new PostgreSQL database - Note the connection string 2. **Deploy the Application** - Connect your GitHub repository to Render - Use the provided `render.yaml` configuration - Set environment variables: - `DATABASE_URL`: Your PostgreSQL connection string - Other variables are pre-configured 3. **Run Database Migrations** ```bash # After deployment, run migrations via Render shell python -m alembic upgrade head ``` ### Environment Variables Required: - `DATABASE_URL`: PostgreSQL connection string - `WANIKANI_API_BASE_URL`: https://api.wanikani.com/v2 Optional (with defaults): - `DEBUG`: false - `LOG_LEVEL`: INFO - `SYNC_INTERVAL_MINUTES`: 30 - `MAX_CONCURRENT_SYNCS`: 3 - `WANIKANI_RATE_LIMIT`: 60 ### Health Check The application exposes a health check endpoint at `/health` for monitoring. ### Monitoring - Use Render's built-in logging and metrics - Application logs include sync status and error information - Consider adding Sentry for error tracking (set `SENTRY_DSN`) ## Local Development with Podman ```bash # Build the container podman build -t wanikani-mcp . # Run with environment file podman run -p 8000:8000 --env-file .env wanikani-mcp # Or run HTTP mode podman run -p 8000:8000 -e DATABASE_URL="sqlite:///./wanikani_mcp.db" wanikani-mcp ``` ## MCP Integration Once deployed, users can connect via: - HTTP mode: Use the deployed URL - Stdio mode: Run locally with `task dev`