sample-first data packs
Every pack previews right on the page: real rows, documented schema, no email wall. Scripts and agents can skip the browser and read catalog.json instead. Full packs ship through Getly. Comparing lookup options? Read HUD fair market rent data, explained and the FMR lookup guide: one at a time or in bulk, plus SAFMR vs FMR for the ZIP-level version. Torn between a file and a query service? CSV vs API for data is a decision table with a real worked example. Hunting for a no-key open data API? Free sample endpoints plus an MCP query layer lists what you can call today. Working with NFL data? See how free nflverse data compares to a preprocessed pack, or start from free sports data CSV: verified sources and licenses, or check the public domain datasets and license table before you build on the rows. Prefer a spreadsheet? Import the NFL games CSV into Excel or Google Sheets walks the 22-row sample step by step. Agents that want raw rows instead can start from the free API datasets page, or wire the catalog straight into Claude Desktop or pi with the MCP server for datasets. Reading someone else's CSV and stuck on column codes? What a data dictionary is breaks down the format with real entries from the NFL and HUD packs. Curious what the llms.txt for AI agents file at this site's root does? The guide breaks down the format with a live example, what is llms.txt covers the definition and how agents use the file, and llms.txt vs sitemap.xml explains how it differs from the sitemap next to it. Building the collection pipeline yourself? The web scraping scripts pack ships the ten Playwright templates we use, and the web scraping tools comparison lays out cloud platforms like Apify against self-hosted scripts. Never written a scraper? Scrape without coding shows the edit-one-config-block workflow behind those templates. Need a few rows fast for a fixture or a demo? The sample data downloads page lists every free sample file with its columns and row count. Want the full story? How we sell public datasets documents the pipeline end to end, license checks to checkout. Running Claude Desktop? Install the dataset MCP server from a config JSON, then compare three free MCP servers for Claude and the rest of the free agent toolkit. Building your own llms.txt? Create one without a generator, using this site's file as the live example. For citations, the NFL, HUD FMR, and Airbnb packs each carry a permanent Zenodo DOI: 10.5281/zenodo.22643177, 10.5281/zenodo.22643149, and 10.5281/zenodo.22643189.
5 datasets shown
Every NFL game since 1999 with scores, spreads, totals, moneylines, and stadium weather.
FY2026 fair market rents for 51,895 ZIP codes, five bedroom sizes per row.
90,169 listings across 6 US cities in one schema, host data stripped.
28 live Superteam Earn bounties with full descriptions, plus 186+ mapped API routes.
10 self-contained Playwright scripts for the scraping patterns you keep rewriting.
120 kindergarten worksheet pages plus 5 Web3 tracking spreadsheets, aggregated into one download.
Every free printable in the funnel on one page: 223 cards grouped by topic, each opening the live page.