CSV & JSON tools that never upload your data
Convert, view, clean, merge, split and validate CSV, JSON, Excel, YAML and XML right in your browser. 28 free tools, no signup, nothing leaves your machine.
Every tool
Convert
- CSV → JSONRows become objects, arrays or JSON Lines.
- JSON → CSVFlatten nested JSON into a spreadsheet.
- YAML → JSONConfig files to JSON.
- JSON → YAMLJSON to readable YAML.
- XML → CSVRepeating XML elements to rows.
- CSV → XMLRows to XML elements or attributes.
- CSV → SQLCREATE TABLE and INSERT statements.
- TSV → CSVTab-separated to comma-separated.
- CSV → TSVComma-separated to tab-separated.
- CSV → MarkdownTables for READMEs and docs.
- CSV → HTMLAccessible HTML table markup.
- CSV → YAMLRows to a YAML list.
- CSV → PDFPrint-ready PDF tables.
- XML → JSONXML documents to JSON.
- JSON → XMLJSON to well-formed XML.
- JSON → TSVJSON to tab-separated values.
Excel
Clean & combine
Why a browser-only toolkit
Most online converters upload your file to a server, process it there and hand back a download. That is a problem when the CSV is a customer list, a payroll export or anything covered by GDPR or an NDA. DataKit does the same jobs with JavaScript in the page itself. Once the page has loaded you can go offline and every tool keeps working.
The tools are built for the formats developers and analysts actually meet: comma, semicolon, tab and pipe-separated files, JSON arrays and JSON Lines, nested API responses, Excel workbooks with several sheets, YAML configs and XML feeds.
Careful with your data
Leading zeros in ZIP codes and IDs are kept. Long numbers are not rounded into scientific notation. Quoted fields with commas and line breaks are parsed by the RFC 4180 rules. Where a tool has to guess (types, delimiters), it tells you and lets you override.
Why some CSV fields get quotes
CSV looks trivial until a value contains the separator itself. Type a value, or pick one, and watch how a correct writer (RFC 4180) has to encode it, and how a naive “join with commas” breaks the row.
- Contains a comma → must be wrapped in quotes
- Contains a double quote → wrap in quotes and write each " as ""
- Contains a line break → must be wrapped in quotes
- Starts or ends with a space → quoting is optional but keeps the space safe
1,Ada,"Arlington, VA",UK
Parsed back: 4 fields ✓ — city = "Arlington, VA"
1,Ada,Arlington, VA,UK
Broken: parsed as 5 fields, so every column after the city shifts right.
Most “my CSV is broken” problems come down to the rule above: a value contains the delimiter and nobody quoted it. If watching something work helps it click for you, the explainers on ahaboo let you manipulate each idea until it makes sense.
Questions
Is DataKit really free?
Yes. There is no account, no trial and no file-size tier. The tools run on your computer, so they cost us almost nothing to offer.
Are my files uploaded?
No. Files are read with the browser’s File API and processed in the page. The only feature that sends data anywhere is the optional “Suggest a fix with AI” button on the validators, and only when you click it.
How large a file can I use?
It depends on your device’s memory. Tens of megabytes are comfortable on a laptop. For very large CSVs, split them first with the CSV splitter.
What is the difference between CSV and JSON?
CSV is a flat table: one record per line, fields separated by commas, no types. JSON is a tree of objects and arrays with numbers, booleans and nulls. CSV suits spreadsheets and bulk imports; JSON suits APIs and nested data.
Does it work on a phone?
Yes, every tool works on mobile browsers, though large tables are easier to handle on a bigger screen.