Edit CSV files like code, not like a spreadsheet.

Trusty CSV opens the file itself and shows it as a table you can read: every column in its own colour, lined up as you type. Edit it with the keys of a code editor, let the quality check find what’s wrong, and clean it with tools made for tables. Nothing is converted and nothing is uploaded: what you save is your file, with only your changes.

freeno sign-upnothing uploaded

coffee-menu.csv
Delimiter: Comma · Header row

why

Code editors are made for code. Spreadsheets, for spreadsheets. Trusty CSV is made for CSV.

A CSV file is a table written as text, and the tools you already have see one half or the other. A spreadsheet sees the table, and converts the text to get it. A code editor keeps the text, and never sees the table.

Trusty CSV sees both. It shows the text as a table you can read, edits it like code, and cleans it with tools made for tables. And it stops there: no formulas, no charts, no fonts, no extensions to set up.

opened-in.csv
opened inwhat you get
a spreadsheeta converted copy: 0042 becomes 42, SEPT2 a date, a long number 9.40011E+21
a code editorlines of commas: sort them, and a row with a line break in a cell falls apart
a data-prep toola project: you import the file, clean a copy and export a new one
Trusty CSVthe file itself, as a table: clean it, save it, and only your changes are new

quality check

See what’s wrong with a file, and fix it in a click

A CSV has usually been through a few hands before it reaches you: exported, pasted together, opened in a spreadsheet and saved again. Quality Check reads the whole file and lists what went wrong on the way. Each kind of problem has a button that fixes every case at once.

customers.csv

28 checks, and only the sure ones

It reports what is surely wrong, or surely written in two ways. It doesn’t guess what your data means.

The list comes in the order to fix things in: first what makes the file read wrong, then the text to clean, then the columns written in two ways. Start at the top and go down. Every fix is one step of Ctrl+Z.

F8 goes to the next problem, and Ctrl+. fixes the one under the cursor. Like everything else, it happens on your computer.

Every check, in the User Guide

checks.csv
problemfix
Rows cut by a line breakJoin them
Rows with delimiters missingAdd them
Empty and repeated rowsRemove them
Broken encoding: EspañaRepair it
HTML entities: Tom & JerryDecode them
Spaces around a valueTrim them
NA and n/a and - in one columnEmpty them
Codes that lost their zeros: 8001Add the zeros
A decimal comma among decimal pointsWrite it like the rest
Dates written in several waysReformat them
Madrid and madrid and MADRIDGroup them
Text in a column of numbersSelect it
Repeated idsSelect them

prepare

From a messy export to the table you need

Group the values written in several ways, filter the rows, rewrite dates and numbers one way, split and join columns, turn columns into rows and back, summarize, and put files together. Each operation changes the file you have open, not a copy of it, and each one is a single Ctrl+Z.

Values written in several ways, made one

Madrid, madrid and MADRID are the same city. Group Similar Values finds them, from the strictest way of comparing to the loosest (same words, same letters, similar spelling, same sound), and turns each group into the value you choose.

Grouping similar values

orders.csv

Filter, and see what was left out

Filter a column by its values, by a condition or by dates. The rows that pass stay where they are, and the rest wait in grey after the last row, so you always see what the filter took away. Operations work on the rows you can see.

Filtering rows

shipments.csv

Columns into rows, and back

A column for each month is easy to read and hard to analyse. Columns to Rows turns them into two columns, a name and a value; Rows to Columns goes back, and Summarize counts, adds up or averages each group. The first rows of the result show before anything changes.

Reshaping the table

stores.csv

And the rest of the cleaning

Each operation takes the selection: the column of the cursor, the cells you selected, or the whole file with Ctrl+A. Quotes, line breaks inside cells and the header row are taken care of, and a toast says how many cells or rows changed, even out of sight.

How the operations work, in the User Guide

operations.csv
menuwhat it does
ColumnsSort by one column or several, as numbers or dates when that’s what they hold
ColumnsSplit a column, join two, or merge the first value that isn’t empty
ColumnsInsert, duplicate, move and delete columns; tidy the header names
RowsRemove the title and notes above the table
CellsReformat numbers and dates, written one way
CellsTrim spaces, change case, remove accents and invisible characters
CellsFill down, fill empty cells, empty the NA, n/a and -
RowsRemove duplicate and empty rows
RowsSplit the values of a cell into rows, and combine them back
TableSummarize: count, add up or average each group
TableTurn columns into rows and back, or swap rows and columns
TableAppend the rows of other files, or join two files by a column
ViewA column’s profile: its values, their counts, a histogram of its numbers or dates
ViewCompare with the last change, the file on disk or another file

editing

A text editor that knows where the columns are

For the edits no menu has: the keys of a code editor, in an editor that tells rows from lines and columns from commas.

Many values, changed at once

Retype every “United States” of a column as “USA” in one go. Select one and press Ctrl+D to add the next in the same column, or Ctrl+Shift+L for all of them, then type. Ctrl+click adds a cursor anywhere.

people.csv

A whole column at once

Retype the dates of a column, fix a prefix or clear a field in thousands of rows: Ctrl+Alt+Shift+C puts a cursor on every value of the column you’re in. Alt+Shift+drag selects a box of columns.

people.csv

Long values stay out of the way

Every row stays one aligned line: values wider than their column fold into “…”, and cells with line breaks into their first line. Click “…”, or move the cursor in, to see the whole value; it folds again when you leave. Ctrl+Alt+E opens the whole column.

coffee-menu.csv

It reads your file for you

Open a file and start: comma, semicolon, tab or pipe; with or without a header row; UTF-8 or Windows-1252; Windows or Unix line breaks. Trusty CSV detects all of it, and keeps it when you save, so the file still opens the same way everywhere else.

ventas-marzo.csv

A real app on your desktop

Trusty CSV for Windows, macOS and Linux is on its way: double-click a .csv file to open it, press Ctrl+S to save straight back to disk, and keep working offline. Tabs you leave open are still there next time.

In the browser, Chrome and Edge save straight back to the original file too; elsewhere, files are saved as downloads.

shortcuts

Shortcuts your fingers already know

The same keys as VS Code, plus a few for working with columns. Press Ctrl+K in the editor for the full list, or see every shortcut in the User Guide.

shortcuts.csv
keysaction
Ctrl+DAdd next occurrence in the same column
Ctrl+Shift+LSelect all occurrences in the column
Ctrl+Alt+DAdd next occurrence anywhere in the file
Ctrl+clickAdd a cursor
Ctrl+Alt+↑ / ↓Add cursor above / below
Ctrl+Alt+CSelect column
Alt+Shift+dragBox selection
Ctrl+F / Ctrl+HFind / replace in a column or the whole file, with regular expressions
Ctrl+GGo to line or row
F8Go to the next problem the quality check found
Ctrl+.Fix the problem under the cursor
Shift+wheelScroll sideways (over the scroll bar, the wheel alone does it)
Ctrl+Shift+PCommand palette
Alt+↑ / ↓Move row up / down, multiline cells included
Alt+← / →Move column left / right
Ctrl+SSave
Ctrl+ZUndo, as many times as you need

trust

The table is the file.

What you see is the text of your file, coloured and lined up on screen. Nothing is converted when it opens and nothing is added to line it up, and saving writes back the same delimiter, quotes, encoding and line breaks. Trusty CSV won’t move a comma you didn’t.

Nothing converted on the way in

Open a CSV in a spreadsheet and it reads each value as it sees fit: codes lose their leading zeros, some turn into dates and long numbers are rounded. Save it, and that’s what the file says from then on.

Trusty CSV reads the text as it is, and writes it back the same.

These are a spreadsheet’s conversions by default; recent versions of Excel let you turn some of them off.

orders.csv, opened in a spreadsheet and saved
idzipskuorderedtracking
4280012-Sep3/4/20269.40011E+21
43280131-Dec3/5/20269.40011E+21
orders.csv, opened in Trusty CSV and saved
idzipskuorderedtracking
004208001SEPT22026-03-049400111202555842332669
004328013DEC12026-03-059400111202555842332676

The diff is what you changed

Repair a broken encoding with the quality check and save: version control sees one line changed, the one you fixed. The rest of the file is byte for byte what it was.

git diff customers.csv
@@ -3 +3 @@
-1042,José Ruiz,Málaga,2026-01-09,free,NA
+1042,José Ruiz,Málaga,2026-01-09,free,NA
  • Nothing is uploaded. Files are read and written on your device. There is no account, and the tabs you leave open stay in your browser’s own storage, so a reload doesn’t lose your work.
  • Nothing changes that you didn’t change. The alignment is only on screen, and an operation rewrites only the cells it is for.
  • Every change is one Ctrl+Z. An operation on a million cells too, and still after you reload the page.
  • The checks don’t guess. Quality Check reports what is surely wrong, and leaves what your data means to you.
Open the editor

faq

Questions

Is Trusty CSV free?

Yes. Open the editor and start working: no sign-up and no limits on the number of files.

Why not just use Excel, or a code editor?

A spreadsheet converts a CSV when it opens it: by default, codes lose their leading zeros, some become dates and long numbers are rounded, and that’s what gets saved. A code editor keeps the text but doesn’t know it’s a table: it can’t line up the columns, sort the rows without breaking a cell that has a line break, or tell you that a row is short. Trusty CSV keeps the text and knows the table. More in Edit CSV files without Excel changing them.

How is it different from OpenRefine or Tableau Prep?

They import your data into a project or a flow, and you export a new file at the end. Trusty CSV opens the file itself, in your browser with nothing to install, shows every row, finds the problems for you, and saves the file with only your changes. It groups similar values, filters, reshapes and joins files too. What it doesn’t do yet is replay a list of steps on the next file, and it reads files, not databases. Compared in detail: an OpenRefine alternative, a Tableau Prep alternative.

Are my files uploaded anywhere?

No. Everything happens in your browser: opening, editing, converting to and from Excel and saving. The files never travel to a server.

Can it find what’s wrong with a CSV file?

Yes. Table › Quality Check reads the whole file and lists its problems: rows cut in two or too short, repeated rows, text with a broken encoding, stray spaces, lost leading zeros, dates, numbers, yes/no values and countries written in several ways, and more, 32 kinds in all. Most have a button that fixes every case at once, and Ctrl+Z undoes it. See Quality Check in the User Guide.

Does the alignment change my file?

No. Columns are aligned only on screen. The file keeps its exact text (the same delimiter, quotes, encoding and line breaks) unless you edit it or choose another format when saving.

How large a file can it open?

Files of tens of megabytes and around a million lines open in a few seconds and stay smooth to edit. Beyond that it depends on your computer’s memory.

Can I open and save Excel files?

Yes: .xlsx, .xls and .ods open sheet by sheet as CSV, and any file can be downloaded as a new .xlsx or .ods workbook. The original workbook is never overwritten, because formulas and formatting can’t be kept in CSV.

Which browsers does it work in?

Any modern browser. Saving straight back to the original file needs Chrome or Edge on a computer; elsewhere, files are saved as downloads.

Is there a desktop app?

It’s on its way, for Windows, macOS and Linux: it opens .csv files with a double-click, saves straight back to disk and works without a connection.