Tools Powered by DuckDB

Kepler.gl is a powerful open-source geospatial analysis tool for large-scale data sets, now embeds duckdb wasm to create geospatial layers.

Generate reports using SQL and markdown. The DuckDB connector allows querying across DuckDB, CSV, Parquet and JSON. (Evidence Cloud)

A tool that provides interactive visualizations for large embeddings. Uses DuckDB.

A next-generation data transformation and modeling framework with support for DuckDB connections for state, transformations & running unit tests locally. (Tobiko Cloud)

A unified SQL query interface and portable runtime to locally materialize (using an embedded DuckDB), accelerate, and query datasets from any database, data warehouse, or data lake. (Spice.ai Cloud)

Tool for effortlessly transforming data sets into powerful, opinionated dashboards using SQL. (Rill Cloud)

Local-first visual ETL/ELT studio. Drag sources, transforms and sinks onto a canvas; it compiles to plain DuckDB SQL and runs entirely on DuckDB. Open source desktop app, with a built-in MCP server for generating and running pipelines from natural language.

Open-source SQL-driven data dashboards, powering Taleshape, built on DuckDB. (managed hosting)

Self-hosted China A-share market data lake; DuckDB views over its Parquet datasets back SQL queries from the CLI and a read-only `run_sql` tool in its MCP server for AI agents.

DuckDB can be used as a caching layer or a data connector in VulcanSQL, a Data API framework for data folks to create REST APIs by writing SQL templates.

Blazing-fast & intuitive pivot tables on Parquet, CSV, JSON files and DuckDB tables in the browser based on DuckDB-Wasm. open-source (MIT). Zero install!

An open-source SIEM for instant log insights. (commercial license)

Minimal RFC 5424 syslog collector and viewer based on DuckDB. Runs as a single, resource-friendly process.

Plug and play analytics for Phoenix applications, powered by DuckDB.

Serverless OLAP API/UI built on top of DuckDB with basic ClickHouse API compatibility and MotherDuck support.

The smallest DuckDB SQL orchestrator on Earth.

An implementation of Snowflake API, enables running queries on Snowflake tables locally with DuckDB without a running warehouse.

Preview csv/tsv, json, and Parquet files in the yazi file manager using duckdb. View the raw data, or a "summarized" view with data-types, min, max, avg etc. for all columns.

Load, explore, transform your datasets and expose them via API. Integration with external APIs, S3, PostgreSQL and ChatGPT.

Agent skills for profiling DuckDB databases, verifying joins, and mapping schemas through a read-only engine, with a seeded local DuckDB demo.

Open-source, privacy-first web analytics for traffic, funnels, ecommerce, Search Console, and AI visibility. Runs as a single Go binary with embedded DuckDB. (HitKeep Cloud)

Fast, accurate, open-source geocoding in Python, using DuckDB.

Open-source data quality tool that profiles tables and generates and runs data quality tests; connects to DuckDB databases and uses DuckDB to profile and test CSV, Parquet, and JSON files locally, in Amazon S3, or in Azure Blob Storage. Apache 2.0.

Business Intelligence Done Right - ReportBurster uses DuckDB as its in-process analytics database engine, and for larger workloads, supports ClickHouse too.

Keeps MySQL tables as Parquet snapshots updated from the binlog, with a generated `views.sql` so DuckDB queries them as tables. Also keeps every version of every row. (pricing)

DuckDB-powered ETL tool written in Go, inspired by evidence.dev's syntax. It uses a structured Markdown config where heading levels define nested blocks, yaml code blocks specify metadata, and sql code blocks handle data interactions. Enables clean, code-light orchestration with minimal setup.

An data mesh platform and high-performance GraphQL backend powered by DuckDB.

No code tool built on top of DuckDB-Wasm and Pyodide that helps build pivot tables from databases of any size with a few clicks.

Play the New York Times Connections Puzzle with DuckDB.

Local, contract-governed data explorer. Federates files and databases (Postgres, Snowflake, BigQuery, Excel, and more) through DuckDB — SQL editor, charts, profiling — then hands AI agents a PII-masked, read-only query surface over MCP. Apache-2.0.

A command line tool to efficiently show end-of-life dates for a number of products in your terminal using the `endoflife.date` API, makes it possible to export the whole `endoflife.date` database as a fully featured DuckDB file.

YAML-based data pipeline framework that runs both locally and fully in-browser designed for data engineers, ML teams, and SaaS developers who need flexible, SQL-powered pipelines.

Virtual warehouse over cloud Parquet. SQL shell, Jupyter/Marimo notebooks, AI natural language queries, and local cache — all powered by DuckDB.

Config-driven geospatial ETL: a YAML pipeline (or visual web editor) of sources, spatial/attribute joins, geoprocessing and field mapping compiles to a chain of DuckDB spatial SQL views and writes GeoPackage or GeoParquet. Reads GeoJSON, Shapefile, FileGDB, WFS, OGC API Features, ArcGIS REST, Parquet, PostgreSQL and more.

Self-hosted analytics stack with a semantic model, DAX-inspired queries compiled to DuckDB SQL, DuckLake storage and dashboards.

Open-source, DuckDB-powered BI workspace for AI-assisted analysis, SQL, charts, and dashboards.

Single-binary backend pairing PocketBase (auth, files, realtime) with an embedded DuckDB runtime: read-only analytics API, cross-source joins over Parquet/CSV/SQLite/PostgreSQL/MySQL, and exports to Parquet, PostgreSQL and Excel.