Free text tools

Why text breaks when you paste it

Because what you copied is not what you saw. Characters that render as a space, a quote or a dash are often something else underneath, and the system you paste into treats them literally.

A non-breaking space is not a space to a form validator. A curly apostrophe is not an apostrophe to a CSV parser. A zero-width space is nothing at all on screen and still counts against a character limit. None of these produce an error message. You get a failed import, a search that finds nothing, or a post that quietly sits over the limit.

Where these characters come from

Four sources, and you have probably used all of them today.

Language models. Model output is typographically polished by default: em dashes, curly quotes, ellipsis as a single character. That polish is exactly what makes it recognisable, and what breaks a plain-text field.

Word processors. Word and Google Docs autocorrect straight quotes into curly ones as you type, and insert non-breaking spaces around punctuation.

Web pages. Copying from a rendered page picks up whatever the CSS and the CMS put there, including the zero-width spaces some sites use to control line wrapping.

PDFs. A PDF has no paragraphs, only lines positioned on a page. Copying one gives you a line break at the end of every visual line, which is why pasted PDF text arrives shredded into fragments.

Which tool to use

Start with the AI text cleaner if you just want the text to work. It fixes the whole set of problem characters in one pass and is the right default when you do not want to think about it.

Reach for something specific once you know what you are after.

Something is wrong and you cannot see what. The invisible character detector shows every hidden character with a line and column. If the accents came out as gibberish, that is a different fault and the encoding fixer is the one for it.

Something in particular is bothering you. The em dash remover, the smart quote converter or the markdown stripper, each doing one job and nothing else.

The text is shredded. A PDF paste needs the line break remover, then usually the whitespace cleaner.

It is a list rather than prose. Strip the markers with the list cleaner, take out repeats with the duplicate line remover, order it with the line sorter, then wrap or reshape it with the prefix and suffix adder or column to comma.

You want to know something about the text. The word counter for length and reading time, the readability checker for how hard it is, the keyword density checker for what it is about, and the text diff for what changed between two versions.

You are converting rather than cleaning. The case converter shows all nine forms at once, the slug generator turns titles into URLs, and the HTML entity and URL tools handle the two escaping schemes people mix up. Everything else is in the grid above.

Nothing is uploaded, and that is not a policy

It is how the tools are built. There is no server to send your text to. The code runs in the page you are already looking at, which is why there is no account, no limit and no queue.

That matters more here than in most categories. Text you want to clean up is often a draft, a client document or something under an agreement. The safest place for it is the machine it is already on.