Charts with DuckDB and Plotly
## Visualization DuckDB MCP Server: In-Process Analytics The **Visualization DuckDB MCP Server** combines DuckDB's high-performance analytics engine with visualization capabilities in Google Antigravity, enabling developers to analyze and visualize data directly without external database dependencies. ### Why Visualization DuckDB MCP? - **Zero Setup**: No database server required - runs in-process with your application - **SQL Analytics**: Full SQL support with advanced analytical functions - **Direct File Queries**: Query CSV, Parquet, JSON files directly without import - **Instant Visualizations**: Generate charts and dashboards from query results - **Blazing Fast**: Vectorized query engine optimized for analytical workloads ### Key Features #### 1. Direct Data Analysis ```python # Query files directly without import result = await duckdb.query(""" SELECT product_category, COUNT(*) as orders, SUM(revenue) as total_revenue, AVG(revenue) as avg_order_value FROM read_parquet('s3://bucket/orders/*.parquet') WHERE order_date >= '2024-01-01' GROUP BY product_category ORDER BY total_revenue DESC """) # Query multiple file formats together combined = await duckdb.query(""" SELECT c.name, SUM(o.amount) as total FROM read_csv('customers.csv') c JOIN read_json('orders.json') o ON c.id = o.customer_id GROUP BY c.name """) ``` #### 2. Visualization Generation ```python # Generate chart from query chart = await duckdb.visualize({ "query": "SELECT month, revenue FROM monthly_sales", "type": "line", "x": "month", "y": "revenue", "title": "Monthly Revenue Trend" }) # Create dashboard with multiple charts dashboard = await duckdb.createDashboard({ "name": "Sales Overview", "charts": [ {"query": revenue_query, "type": "line"}, {"query": category_query, "type": "bar"}, {"query": region_query, "type": "pie"} ] }) ``` ### Configuration ```json { "mcpServers": { "visualization-duckdb": { "command": "npx", "args": ["-y", "@anthropic/mcp-visualization-duckdb"], "env": { "DUCKDB_MEMORY_LIMIT": "4GB", "DUCKDB_THREADS": "4", "AWS_ACCESS_KEY_ID": "your-key", "AWS_SECRET_ACCESS_KEY": "your-secret" } } } } ``` ### Use Cases **Data Exploration**: Quickly explore and visualize datasets without database setup or data import. **Local Analytics**: Run complex analytics on local files during development and testing. **Prototype Dashboards**: Rapidly prototype data visualizations before building production solutions. The Visualization DuckDB MCP Server provides instant analytics and visualization capabilities for any data.
{
"mcpServers": {
"visualization-duckdb": {
"mcpServers": {
"visualization": {
"args": [
"-y",
"@xoniks/mcp-visualization-duckdb"
],
"command": "npx"
}
}
}
}
}