Query data where you code. Fast in-memory analytics with DuckDB inside VS Code.
Product Page | Roadmap | Contributing | PlugMonkey
Stop context-switching. Query Parquet, CSV, and JSON files directly in VS Code — no imports, no database setup, no leaving your editor.
| Before | After |
|---|---|
| Open terminal → Write Python → Read file → Print results | Right-click file → View Data |
| Install database → Configure connection → Switch apps | Install extension → Query |
| 5-10 minutes | 5 seconds |
- Install the extension from VS Code Marketplace
- Connect via the DuckDB sidebar (Cmd+Shift+P → "DuckDB: Create Connection")
- Query any SQL file with Cmd+Enter
Or just right-click any Parquet/CSV/JSON file → "View Data" to explore instantly.
Query data files directly without any import step:
-- Query a Parquet file
SELECT * FROM 'data/sales.parquet' WHERE amount > 1000;
-- Join multiple CSV files
SELECT a.*, b.category FROM 'orders.csv' a JOIN 'products.csv' b ON a.product_id = b.id;
-- Aggregate JSON data
SELECT date, SUM(value) FROM 'events.json' GROUP BY date;Supported formats: Parquet, CSV, TSV, JSON, JSONL, NDJSON
| Source | Description |
|---|---|
| In-Memory | Fast, temporary database for scratch work |
| File (.duckdb) | Persistent local database |
| MotherDuck | Cloud data warehouse |
| PostgreSQL | Query external Postgres databases |
| S3 | Query Parquet/CSV in your data lake |
Browse your data visually:
- Databases → Schemas → Tables → Columns
- See column types at a glance
- Row counts per table
- Right-click context menus for quick actions
Modern data grid with:
- Sortable columns
- Pagination for large results
- Export to CSV, JSON, TSV
- Query history
- Copy cells/rows to clipboard
- Autocomplete for tables, columns, keywords
- Snippets — type
csv+ Tab,parquet+ Tab,summarize+ Tab - Explain query — see execution plans
- Cancel queries — stop long-running operations
| Command | Windows/Linux | Mac |
|---|---|---|
| Run Query | Ctrl+Enter |
Cmd+Enter |
| Run Selected | Ctrl+Shift+Enter |
Cmd+Shift+Enter |
| Explain Query | Ctrl+Shift+E |
Cmd+Shift+E |
| New SQL File | Ctrl+Alt+N |
Cmd+Alt+N |
- Explore CSV/Parquet files instantly
- Filter and export subsets of data
- Get quick statistics with
SUMMARIZE
- Test ETL queries locally
- Query S3 data lakes
- Federated queries to PostgreSQL
- Prototype database schemas
- Test SQL before production
- Stay in your IDE
- 10-100x faster than Pandas for large files
- SQL alternative for quick EDA
- Export results for notebooks
Type prefix + Tab to expand:
| Prefix | Expands To |
|---|---|
csv |
SELECT * FROM read_csv('...') |
parquet |
SELECT * FROM read_parquet('...') |
json |
SELECT * FROM read_json('...') |
summarize |
SUMMARIZE table_name |
pivot |
DuckDB PIVOT syntax |
ctas |
CREATE TABLE AS SELECT |
| Setting | Default | Description |
|---|---|---|
maxResultRows |
10,000 | Max rows per query |
defaultResultLimit |
100 | Default LIMIT for previews |
queryTimeout |
0 | Timeout in seconds (0 = none) |
autoConnect |
false | Auto-connect on startup |
s3.defaultRegion |
us-east-1 | Default AWS region |
postgres.defaultPort |
5432 | Default Postgres port |
- View Data / New Query / Show Info
- Copy SELECT / CREATE TABLE / Table Name
- Copy as CSV / JSON
- Export / Rename / Truncate / Drop
- Filter by Column → generates WHERE query
- Sort by Column → generates ORDER BY query
- Insert / Copy Column Name
- View Data / Describe Schema
- Import to Table
- New Query / Copy SELECT
- VS Code 1.85.0+
- No external dependencies (DuckDB bundled)
- Core query execution
- Schema explorer
- File querying (Parquet, CSV, JSON)
- Remote connections (MotherDuck, PostgreSQL, S3)
- Query history panel
- Basic visualizations
- AI-powered SQL (Premium)
See ROADMAP.md for the full roadmap.
Contributions welcome! See CONTRIBUTING.md for guidelines.
Built with DuckDB — the fast in-process analytical database.
PlugMonkey builds developer tools and VS Code extensions to supercharge your workflow.
MIT — Built by PlugMonkey