dataaaaa!

Streambed is a CDC engine that offloads analytical queries from Postgres to Apache Iceberg without Spark or ETL. It decodes WAL messages, writes Parquet to S3, and uses copy-on-write for updates.

ktx is a self-improving context layer that automates semantic mapping and wiki ingestion for data agents. It resolves fan traps and detects joinable columns to ensure LLMs use canonical SQL. ktx ingests those layers and combines them with raw-table introspection and wiki content.

Quacklake is a Cloudflare Workers service implementing DuckDB's Quack protocol. It uses Durable Objects with SQLite to manage DuckLake catalog metadata and R2 for storage. It is useful for protocol integration work, local Worker tests, and R2-backed DuckLake metadata smoke tests.

Duckle is a visual data pipeline studio that compiles drag-and-drop graphs into SQL for execution via DuckDB. It features a local AI assistant for pipeline generation and supports over 290 connectors. Duckle compiles the graph to SQL and executes it through a real columnar engine.

Gribouille is a Typst library implementing the Grammar of Graphics via a declarative API. It enables data engineers to build complex visualizations using layers, scales, and aesthetics.

Siterows enables querying websites using SQL to extract structured data without writing custom scraping code. Users can automate the process via API. Query any website with SQL ... no scraping code required. This tool simplifies data ingestion by treating web pages as relational tables.

NASA's General Mission Analysis Tool is an open-source enterprise system for space mission design and navigation. It supports diverse flight regimes including low Earth orbit and deep space.

This marketplace provides plugins and scripts to enable agentic development for Power BI and Fabric. It allows tools like Claude Code to manage semantic models and deployment pipelines. Agentic development is when you use agents to help you build, manage, and optimize artifacts and software.

DuckLake SDK provides native Rust and Python interfaces to manage DuckLake tables without requiring DuckDB. It handles metadata via SQL and data as Parquet. All language SDKs are built on the same Rust core, which bundles the implementation of the DuckLake specification. It supports Polars.

This toolkit provides a high-performance telemetry platform for Dota 2 match state analysis and real-time data visualization. It features position prediction, hidden particle analytics, and automated input sequencing.

Goldenmatch is a polyglot toolkit for data quality and entity resolution, featuring a Rust extension layer for Postgres and DuckDB. It uses a chunked streaming reader to process 5M records in 50 minutes on commodity hardware. Chunked mode now actually delivers on its 1M to 100M+ promise.

pg_duckpipe is a PostgreSQL extension enabling real-time CDC to sync row tables into DuckLake columnar tables for HTAP workloads. It uses logical replication to stream WAL changes with tunable latency. Run transactional and analytical workloads in a single database by using a one-command setup.

Arnio is a compiled C++ data cleaning engine designed to eliminate Python parsing overhead. It performs type inference, normalization, and deduplication in columnar memory before exporting to pandas. Your CSV hits C++ before Python even wakes up. It uses Python only for configuration.

AIPOCH provides 553 specialized agent skills to automate medical research workflows using Claude Code or OpenClaw. It features professional medical research logic for data analysis and protocol design. The MedSkillAudit framework ensures reliability by assessing skill readiness before deployment.

MemMe is a local-first AI memory engine built in Rust that uses a single SQLite file for storage. It features sub-10ms search and knowledge graph integration. Your AI conversation memory sits on OpenAI's servers. That data belongs on your device. It supports six language bindings for easy embedding.

This repository provides a production-ready prompt library for Claude AI and coding agents based on the APEI cycle. It enforces a minimal approach using a core system prompt and specific specialist prompts to ensure each recommendation must end with a concrete decision, tool, or validation step.

html-anything is an agent skill that transforms complex data, URLs, and files into verified single-file HTML artifacts. It uses source-aware parsing and 17 style systems to replace long Markdown replies with interactive pages. The skill figures out the source, chooses the right use case and style.

OpenFoundry is an open-source operational data platform built as a Go monorepo. It replicates the Palantir Foundry capability model, including datasets, ontology, and governance. The architecture features 41 microservices and 32 shared libraries.

The Cortex Code Skill routes Snowflake operations to the Cortex Code CLI, allowing coding agents to use specialized Snowflake expertise in headless mode. It integrates with tools like Claude Code and Cursor to automate data workflows. The skill routes Snowflake-related operations to Cortex Code CLI.

LangPulse provides daily snapshots of GitHub repository activity by language and star count. It uses Go and GitHub Actions to query the Search API, tracking repos pushed to within a 30-day window. LangPulse stores a single integer: the total count returned by a GitHub repository-search query.

The Snowflake AI Kit provides developer tools for building on Snowflake using AI coding agents. It features a one-command installer for the Snowflake and Cortex Code CLIs, plus a Claude Code plugin. The installer sets up Snowflake CLI and Cortex Code CLI to enable 40+ built-in data skills.

DuckDB Raster is a community extension that integrates geospatial raster data into SQL workflows using GDAL. It exposes formats like GeoTIFF and Cloud Optimized GeoTIFF as tables, allowing for tile-based filtering and band math.

Datanomy is a terminal-based tool for inspecting the internal structure and metadata of columnar data files. Built with Textual and PyArrow, it currently supports Parquet files. Datanomy is a terminal-based tool for inspecting and understanding data files. It provides an interactive view.

Honker adds Postgres-style NOTIFY/LISTEN to SQLite, enabling durable pub/sub and task queues without a separate broker. It uses a background thread to poll PRAGMA data_version, achieving sub-millisecond latency. The queue is just rows in a table with a partial index for ACID compliance.

ZON is a token-efficient serialization format designed to replace JSON in LLM workflows. It uses tabular encoding and single-character primitives to achieve 35-70% token reduction. ZON is the only format that wins (or ties for first) on every single LLM while maintaining 100% data fidelity.