FolderDB watches a folder for CSV and Excel exports, auto-maps every column and row type, and loads them into SQLite or Postgres — without writing a single line of ETL.
$ pip install folderdb
$ fdb watch ~/Downloads
[15:42:01] Wandering /home/jasper/Downloads for new files...
[15:42:03] 📄 Detected: leads_export_2026-09-24.csv
[15:42:03] ↳ 12 columns mapped (email, name, company, ...)
[15:42:04] ↳ 1,847 rows → leads_export_2026_09_24 (SQLite)
[15:42:04] Done. Query with: sqlite3 ~/.folderdb/data.db
$ sqlite3 ~/.folderdb/data.db "SELECT COUNT(*) FROM leads_export_2026_09_24"
1847
Set it up in 30 seconds. Never think about imports again.
Point it at any folder. New CSV or Excel files are detected, stabilised (waits for writes to finish), and imported automatically.
Detects column types — integers, floats, dates, booleans, text. Fuzzy-matches column names (e-mail → email) so you get clean tables.
New column appears in your file? FolderDB widens the table automatically. No manual ALTER TABLE. No migration scripts.
Same filename appears again? Cached column mapping means zero reconfiguration. Works great with recurring exports.
Start with zero-config SQLite. Scale up to Postgres by setting DATABASE_URL. Same tool, same workflow, same no-code promise.
Your data never leaves your machine. No cloud sync, no telemetry, no third-party servers. License verification is the only external call.
pip install folderdb and optionally set a target directory and database.
Any CSV or Excel file lands in your watched folder. FolderDB detects it, samples rows, and maps every column.
Connect any SQL tool — DataGrip, Tableau, a Python notebook — and query your data as tables. Done.
One product. One price. No surprise bills.
Weekly CSV exports from Salesforce, HubSpot, or your CRM land in a folder. FolderDB loads them into Postgres automatically for your BI dashboards.
Finance sends Excel reports by email. Save them to a watched folder instead of manually importing. Your database stays up to date without a ticket.
Survey data, sensor logs, experimental results as CSVs. FolderDB builds your analysis database incrementally — no schema planning needed.
Running a side project on SQLite. Save data exports into a folder and get queryable tables instantly. No ETL tooling, no engineering time.