Repository navigation
Datasette
title: Datasette type: tool created: 2026-10-08 last_updated: 2026-10-08 related: ["Datasette Live-Plus-Static Hosting", "Datasette Plugin Authoring"] sources: ["https://docs.datasette.io/en/stable/", "https://github.com/simonw/datasette"] radar_quadrant: Tools radar_ring: Trial radar_position: inner
Sharing a data file usually means building a website and an API around it first. Datasette removes that step: it publishes data of any shape as an interactive, explorable website with a JSON API, so people can browse, filter and query without anyone writing the serving layer.
The site lets visitors browse tables, filter rows and run saved SQL queries, called canned queries, which can take parameters and can optionally be made writable. A JSON API exposes table and query data with pagination, filtering and foreign key expansion. Built-in commands deploy a site to hosts such as Google Cloud Run, Heroku, Vercel and Fly, or package it as a container with datasette package.
Plugins are installable extensions, and the documentation covers installing, configuring and writing them. Permissions can be set per instance, database, table, canned query and arbitrary SQL execution through configuration, which is how a login-gated live copy can sit alongside a public one.
Datasette is placed in Trial at the inner position, on the owner's confirmed first-person use: it is the exploration and serving layer of the owner's pyholofoil pipeline, with several plugins pinned for editing, saved queries, charts, password login and MCP access. The repository is Apache-2.0 licensed, was created in 2017, and had about 11,500 stars and 689 open issues when checked. It stays below Adopt because the use covers one project and the pipeline still carries open decisions.
