NamerynAI Text Cleaner
Paste text from ChatGPT, Claude or Gemini and get plain text back: no markdown stars, no em dashes, no curly quotes, no emojis, no hidden characters. Nothing leaves your browser.
Clean text
Nothing removed yet.
What the AI Text Cleaner Removes
- Markdown. Heading hashes,
**bold**and*italic*stars, backticks, quote markers, horizontal rules and code fences. Links keep their text and address; star bullets become plain dashes. Code inside a fence stays exactly as written.## Plan → Plan · **ship** it → ship it
- Em dashes. Each em dash, and each spaced en dash, becomes a comma or a period, your choice. A dash in a number range like 1–3 stays.
fast — and cheap → fast, and cheap
- Curly quotes. “Smart” double and single quotes become straight quotes, and the single ellipsis character becomes three dots.
“it’s done…” → "it's done..."
- Emojis. Every emoji, with its skin tone and joined parts, and the space it leaves behind. ©, ® and ™ stay.
Great work 🎉. → Great work.
- Hidden characters. Zero-width spaces and joiners, byte order marks, soft hyphens, direction marks and Unicode tag characters go; non-breaking spaces become ordinary spaces.
Invisible in the box, counted in the report.
- Doubled spaces. Runs of spaces inside a line, spaces at the end of a line and stacks of blank lines. Indentation at the start of a line stays.
done. Next → done. Next
How to Clean AI-Generated Text
- Paste the text from your AI chat into the box.
- Choose what to remove. Every switch is on by default; turn one off to keep that class, for example the emojis in a social post.
- Pick comma or period for the em dashes. A comma keeps the sentence; a period splits it in two and capitalizes the next word.
- Read the count, then Copy. The line under the clean text says what was removed, so a paste with nothing to fix tells you so.
Why AI Text Needs Cleaning
Chat assistants write in markdown because their own windows render it. Pasted into an email, a document or a form, the same text arrives with stray stars, hash signs and backticks. Copying from a chat can also carry characters you cannot see: zero-width spaces, non-breaking spaces and direction marks that break search, word counts and spell checkers, or make two identical-looking words compare as different.
The em dash is a habit of its own. Models lean on it to join clauses far more often than most people do, and readers now notice it. The cleaner swaps each one for the comma or full stop most writers would have used. It never rewrites your words, so the voice of the text is still yours to fix.
Frequently Asked Questions
Does my text leave my browser?
No. The cleaning runs in the page's own JavaScript on your device. The text is not sent to a server, not stored and not logged.
Will this make AI text pass an AI detector?
No, and no tool honestly can. The cleaner removes formatting, punctuation habits and invisible characters; it never changes your words. Detectors judge word choice and sentence structure, so if the writing needs to sound like you, rewrite it in your own words.
What hidden characters does it remove?
Zero-width spaces, zero-width joiners and non-joiners outside emojis, word joiners, the byte order mark, soft hyphens, left-to-right and right-to-left marks and overrides, invisible math operators and Unicode tag characters. Non-breaking and other unusual spaces become ordinary spaces.
Why does ChatGPT use so many em dashes?
Large language models learn from edited prose, where the em dash is common, and they lean on it to join clauses. Most everyday writing uses a comma or a full stop in the same place, which is what the cleaner puts there.
Does it change my words?
No. It only touches formatting marks, dashes, quotes, emojis, invisible characters and spacing. Text inside a code block is left exactly as written.