A folder of CSVs, queryable in minutes

Query and Browse CSV Files Like a Database

DbSchema opens a directory of CSV files through a JDBC driver and treats each file as a table: column types inferred automatically, SQL joins across files, a visual data explorer, and diagrams — no server, no import step.

  • Point DbSchema at a folder — every file becomes a table
  • Column names and types read from headers and sampled rows
  • Join, filter, and aggregate across files with plain SQL
  • Refine inferred columns before migrating anywhere
available for ISO-27001 certified ISO-27001 certified
Free Community Edition No signup required 15-day Architect trial included
Free Download
DbSchema showing a folder of CSV files as diagrammed, queryable tables
100+ Databases
|See All   >

Trusted by teams at

Spreadsheets Stop Scaling
Long Before Your CSV Folder Does

A data delivery lands as forty flat files, and the questions begin: which columns join, which types lie, which file is the fact table. A spreadsheet answers one file at a time, slowly.

DbSchema treats the whole folder as a schema — each file a table, each column typed by inference — so you interrogate it with SQL and diagrams instead of scrolling.

Open, Join & Verify
Flat Files Like Database Tables

Everything a database client offers, applied straight to files on disk.

DbSchema inferring a schema from a directory of CSV files
A Folder In, a Schema Out

DbSchema samples each file, infers column names and types from headers and rows, and places the resulting tables on a diagram canvas — reviewable and adjustable before any migration.

Build SQL joins across CSV files visually in DbSchema
Cross-File Joins by Point and Click

Compose SELECT, JOIN, and aggregation queries over several files in the visual builder, then read and rerun the generated SQL.

Browse CSV rows in the DbSchema data explorer
Inspect Files Too Big for Excel

Filter, sort, and page through rows in the data explorer to check that values match their declared types before an ETL run depends on them.

Hand Over a Data Dictionary
Not Just a Zip of Files

One export builds interactive HTML5 documentation of every inferred table — columns, types, and the annotations you added. Whoever receives the data reads what it contains in a browser before opening a single file.

Explore the Documentation Tool

HTML5 documentation generated from CSV-derived table definitions in DbSchema

How DbSchema Opens a CSV Folder

Two routes lead in. The CsvJdbc driver (csvjdbc.jar) with the URL jdbc:relique:csv:/path/to/directory maps every .csv in that directory to a table; its connection properties tune type inference when the defaults miss. Alternatively, a DuckDB in-memory connection reads files via read_csv_auto('/path/to/file.csv') in the SQL editor. DbSchema also maintains its own open-source CSV JDBC driver on GitHub — setup notes live on the CSV JDBC driver page.

Turn That CSV Folder into a Schema

No server, no account — just the free desktop app, with a 15-day Architect trial inside.

Download Free

Teams working with CSV Files often use these engines too. Explore dedicated guides and JDBC setup for each.

Browse all 100+ supported databases