CSV Validator

How to use

1. 1Paste a CSV into the input area, open a .csv or .tsv file from your device, or click Load valid sample or Load broken sample to see the validator in action.
2. 2Leave the delimiter on Auto to let the validator pick the most likely separator, or pin a specific delimiter (comma, tab, semicolon, or pipe) when you know the format.
3. 3Toggle Header row to choose whether the first row should be treated as column names. Duplicate or empty header checks only run when this is on.
4. 4Read the verdict, the issue count, and the list of findings. Errors block downstream consumers; warnings deserve a second look; info notes are background context.
5. 5If parsing failed, read the line, column, and caret snippet under Fatal parse error to find the exact character that broke the file.
6. 6If row widths are inconsistent, scan the table of rows with the wrong column count and compare actual width against the expected width.
7. 7Use Copy report to save the full diagnostic summary for a code review, an issue, or your own notes.

About this tool

CSV Validator parses any CSV (or TSV, semicolon, or pipe-separated file) in your browser and reports every problem you would want to know about before sending the file downstream. The parser is RFC 4180 compliant: it understands quoted fields, doubled-quote escapes inside quoted fields, embedded line breaks inside quoted fields, and the three standard line endings (LF, CR, and CRLF). When the parser cannot continue, the tool reports the exact line, column, and 0-based character offset of the fatal error, plus a plain English hint that distinguishes the three most common failure modes: an unterminated quoted field, an unexpected quote in the middle of an unquoted field, and a stray character that breaks the row structure. A two-line context snippet with a caret marker shows the offending character so you can fix it without scrolling through the file. The tool also reports the structural issues that parsers ignore but spreadsheets and databases reject. It computes the most common row width and lists every row whose column count differs, with the row index, the source line range, the actual width, and a six-field preview so you can see why it was rejected. It detects mixed line endings (LF, CRLF, and CR mixed in the same file), the presence of a UTF-8 byte order mark at the start of the file, embedded line breaks inside quoted fields that confuse naive line-by-line consumers, duplicate header names and empty header cells when a header row is declared, and trailing blank rows. Delimiter detection runs separately from the active delimiter: the tool scores comma, tab, semicolon, and pipe candidates against the first 32 KB of the file and reports the winner, so you can confirm at a glance that the file matches the format you think it is. Diagnostics surface the active delimiter, the detected delimiter, the total number of physical lines, the number of parsed rows, the maximum column count, and the line ending breakdown. A copy-friendly text report lists every finding so you can paste it into an issue tracker, a pull request review, or an internal note. Useful for data engineers checking a vendor export before it hits a pipeline, analysts confirming a download from a dashboard, developers debugging a failing import, and anyone who has been bitten by a single misplaced quote at row 13,427. Everything runs locally in your browser. The CSV is not uploaded.

Free to use. Works in your browser. No signup, no login.