Browser Kitty
日本語
data

DuckDB Explorer

Open DuckDB files locally and inspect structure, data, guarded read-only SQL, and database differences.

View on GitHub
Local processing
About your filesFiles and content selected in this tool are processed in your browser and are not uploaded to a Browser Kitty backend.
Operation demo See the basic workflow in a short video.

About this tool

DuckDB Explorer is a read-only browser tool for opening a local DuckDB database, understanding its overall structure first, and then drilling into the data you need. Inspect schemas, tables, views, columns, estimated row counts, file size, and DuckDB version, then browse tables and views 100 rows at a time.

It also supports search, column filters, sorting, nested-value details, on-demand column profiling, guarded read-only SQL, data-dictionary HTML export, and structural comparison with another DuckDB file. The app does not send selected database files, SQL text, searches, or results to an external server.

Good for

  • Quickly inspect which schemas, tables, views, and columns are present in a received DuckDB file
  • Investigate data with search, sorting, column profiling, and read-only SQL without modifying the database
  • Compare structural differences such as tables, column definitions, and view SQL between two DuckDB files
  • Save database structure and completed column profiles as a shareable data-dictionary HTML

What it can do

Key capabilities available in this tool.

How to use

  1. Select Open tool and choose or drop a .duckdb or .db file.
  2. Use Overview to inspect schemas, tables, views, column counts, estimated row counts, file size, and other database metadata.
  3. Select a table or view, then use search, filters, sorting, cell details, and the Columns view to inspect its contents.
  4. Run profiling only on columns you need to inspect, including NULL counts, estimated distinct counts, numeric summaries, date ranges, text lengths, and common values.
  5. Use SQL to run one read-only SELECT / WITH / VALUES / SHOW / DESCRIBE / DESC / EXPLAIN statement at a time.
  6. When needed, export a data-dictionary HTML from Overview or choose another DuckDB file for structural comparison.

Supported

  • Input: .duckdb / .db. The selected database is processed by DuckDB-Wasm in the browser and opened read-only.
  • Data browser: pages through 100 rows at a time with search, column filters, sorting, and details for nested LIST / STRUCT / MAP values.
  • Column profiling: NULL counts, estimated distinct counts, numeric summaries, date ranges, text lengths, and common values run only for explicitly selected columns.
  • SQL: runs one SELECT / WITH / VALUES / SHOW / DESCRIBE / DESC / EXPLAIN statement at a time. SELECT-style display/export is capped at 1,000 rows.
  • Exports: viewer CSV / JSON covers the current page only. Data-dictionary HTML includes structure, view definitions, and completed column profiles, but not table rows.
  • Comparison: compares schemas, tables, views, column definitions, view SQL, and estimated row counts; it does not perform row-level data diffs.
  • Browsers: Chrome and Edge are the primary targets. The generated standalone HTML can be opened directly with file:// and used without a network connection.

Limitations and notes

  • The app does not edit or write DuckDB files, and SQL is restricted to read-only statements.
  • CSV / JSON export from the data browser is limited to the currently displayed page.
  • SELECT-style SQL results are limited to 1,000 rows for display and export.
  • Database comparison is structural and does not report row-level inserts, updates, or deletes.
  • Column profiling is not an automatic full-column scan; it runs only for explicitly chosen columns. Data-dictionary reports may include common values from profiles you ran.
  • Databases that depend on incompatible DuckDB versions or optional extensions may not open in the browser build.
  • Large databases, expensive views, searches, profiles, or comparisons can use substantial CPU and memory. Firefox and Safari support is best-effort.

Frequently asked questions

Are DuckDB files uploaded to a server?

No. Selected databases are processed by embedded DuckDB-Wasm in the browser. The app does not send database files, searches, filters, SQL text, query history, or query results to an external server.

Can SQL modify the database?

No. The database is opened read-only, and SQL is limited to one read-only SELECT, WITH, VALUES, SHOW, DESCRIBE / DESC, or EXPLAIN-style statement. Writes, DDL, ATTACH, extension loading, and similar operations are rejected before execution.

Can I export the entire DuckDB database as CSV?

Viewer CSV / JSON export covers the current 100-row page, while SELECT-style SQL output is capped at 1,000 rows. These limits avoid unintentionally expanding very large data sets in browser memory.

Can it compare changed rows between two databases?

No row-by-row diff is performed. The comparison focuses on database structure such as schemas, tables, views, column definitions, view SQL, and estimated row counts.

Does the data-dictionary HTML include table row data?

It does not include table rows. However, common values from column profiles explicitly run in the current session may be included, so review the report before sharing it.

Offline version and source code

The root-level duckdb-explorer.html is a standalone HTML containing the required DuckDB-Wasm runtime. After saving it, open it directly with file:// to browse databases, run guarded SQL and profiles, export data dictionaries, and compare structure without runtime network access.

View source on GitHub

Use together

Useful next steps and companion tools, organized around the workflow.

Inspect SQLite instead SQLite Explorer Use SQLite Explorer when the source is SQLite and you want cross-table search, record relationships, and read-only SQL. Inspect Parquet files Parquet Viewer Use Parquet Viewer when you need schema, metadata, row-group details, and previews for Parquet files rather than a database. Inspect text-based data JSON / YAML / CSV Viewer Use it to view, search, analyze, and convert JSON, YAML, CSV, TSV, and JSON Lines.