Kuery

Databases and files.One query.

Kuery is a native desktop app, written in Rust, for browsing PostgreSQL, MySQL, SQL Server, SQLite, Redis and MongoDB, and for running SQL over CSV, JSON and Parquet without importing anything.

Download for macOS View the source

v0.5.0 for macOS, Windows and Linux. Built with GPUI, Apache Arrow and Polars. Try the editor on the right: pick a source and press Run.

Polars SQL · read only 




ResultsProfile



















Eight sources. One grid.


Database SQL speaks each engine's own dialect. File SQL runs Polars over complete inputs. Every result lands in the same virtualized grid, with the same Profile view. Open a row to see what each source supports.
























Everything after the first query


Kuery is built for the part that comes next: reading the data, explaining it, and changing it without surprises. Each view below is a real capture of the app with synthetic data.












Open the file. Skip the import.


CSV and TSV, JSON arrays, JSONL and Parquet open in the same grid you use for query results, with native or inferred types for every column.


Change the delimiter, header and NULL options, then reload. The grid always says whether you are looking at a preview or the complete file.












Understand the result, not only the rows


Profile shows missing and distinct values, distributions, percentiles, categories and timelines for the rows you just pulled.


Compare groups with box plots, cross tabs, time breakdowns and numeric correlations. Analysis runs when you ask for it, on the result in front of you.












See where the time went


Read EXPLAIN output as a tree or a graph. Every step shows its rows, cost and share of the total time; PostgreSQL adds measured timings, row-estimate checks and plain-language insights.


Compare two plans before you ship the index.












Writes you read before they run


Every project starts read-only. Choose an explicit write mode and Kuery previews the impact, flags an UPDATE with no WHERE, and keeps the changes in a transaction until you commit.


Environment policies decide what development, staging and production allow. Table edits are staged and checked before they are saved.












Browse without writing a query


Connections, schemas, tables, views, routines, triggers and types in one explorer. Filter, sort and page on the server, then inspect columns, keys, indexes and definitions.


Schema diagrams draw foreign keys with crow's-foot relationships and remember where you put each card.












Redis, with the same care


Browse keys with SCAN and inspect strings, JSON, hashes, lists, sets, sorted sets and streams. Run commands in a console built for it.


Read-only by default. Redis writes apply at once and cannot be rolled back, so Kuery makes you choose them on purpose.














From folder to answer






First, open a folder.


A Kuery project is a plain folder with a small .kuery directory: connection definitions, saved SQL and diagram positions. Passwords go to your system's credential store, never into the project.






Then open a file or a connection.


Press ⌘⇧O for CSV, JSON or Parquet, or add PostgreSQL, MySQL, SQL Server, SQLite, Redis or MongoDB with TLS and an OpenSSH tunnel.






Finally, ask it in SQL.


Join files with Polars SQL or query the database in its own dialect. Copy cells and rows as text, TSV or JSON, or export to a file.














Open project⌘O


retail-analytics/
├── orders.csv
├── regional_targets.json
├── events.parquet
└── .kuery/
    ├── project.toml   # connections, no passwords
    ├── queries/       # saved SQL
    └── diagrams/      # card positions










orders.csvDataColumnsProfile618 rows · complete file


order_idi64monthstrregionstrstatusstrrevenuef64


104212026-06Europecompleted342.79


104222026-06Latin Americacompleted399.16


104232026-06North Americarefunded298.50


104242026-06Europecompleted618.32


104252026-05Latin Americacompleted705.84


104262026-05North Americacompleted524.00


104272026-05Europepending76.70










File SQLPolars SQL · read only


SELECT region, COUNT(*) AS orders,
  ROUND(SUM(revenue), 2) AS revenue
FROM orders
WHERE status = 'completed'
GROUP BY region ORDER BY revenue DESC;


regionordersrevenue


Latin America20682226.96


Europe20670713.87


North America20661915.01


3 result rows · 19 ms · full file inputs

















Also in the box


Schema diagramsCrow's-foot


SSH tunnelsOpenSSH


TLSPer driver


PasswordsOS credential store


Export resultsCSV, TSV, JSON


Copy cells and rowsText, TSV, JSON


Environment policiesdev, staging, prod


Query cancellationTimeouts too


Saved SQL and historyPer project


Recent projectsWelcome window






Keys you will use daily




Command palette⌘K


Open project⌘O


Open data file⌘⇧O


Run statement or selection⌘↵


Format SQL⌘⇧F


Cancel query⌘.


Save query⌘S


Toggle sidebar⌘B


Close tab⌘W


Every shortcut is configurable in keymap.toml.












Contributors




    

  • angel-afonso
    • 

  • HectorZR
    • 











Run your first query.


Download Kuery 0.5.0, open a folder, and press ⌘↵. Every release ships with SHA256 checksums. Kuery is an early prototype under active development, so expect it to move fast.








Download Kueryv0.5.0 · latest release












kuery-v0.5.0-universal-apple-darwin.dmg


Universal app for Apple silicon and Intel. Open the disk image and drag Kuery onto Applications.

Download .dmg

All releasesSHA256SUMS