CSVbox

SaaS teams that need to let users upload CSV, Excel, or other spreadsheet files often end up building their own import UI, and what looks like a simple upload form turns into a much bigger project. Handling column mapping, locale-specific date formats, Excel quirks, large files up to 500MB, and server-side validation is described as the other 95% of the work beyond basic parsing. Teams that build this themselves face weeks of development, browsers that choke on big files, edge cases that break in production, and ongoing maintenance sprints. One customer quoted on the site says their self-built version was buggy and hard for users before they switched.

CSVbox is a single embeddable JS widget that adds a complete import experience to any web app: file upload, AI-powered column mapping, inline per-row validation, and delivery of clean data to a webhook, API, database, or 20+ destinations like Postgres, Airtable, Google Sheets, Zapier, and n8n. Setup involves configuring a schema and validation rules in a dashboard, embedding the widget or sharing an Import Link, and then receiving validated data. It supports CSV, Excel, PDF, images, and documents, streams uploads up to 2 million rows without freezing the browser, and offers AI Pre-Import Transform to restructure messy files from a plain-English instruction before mapping. A Private Mode option runs the entire import process in the user's browser so no file or row data reaches CSVbox servers, and an MCP server lets developers configure importers from AI clients like Claude or Cursor.