Negative Prompt Generator
Build a Stable Diffusion negative prompt from the problems you keep getting.
Strip markdown, smart quotes and invisible characters out of generated text.
Each one reports how many characters it changed, so nothing is removed silently.
The AI Text Cleaner strips the formatting that rides along when you copy text out of an AI assistant: markdown syntax, curly quotes and em dashes, emoji, stacked blank lines and the invisible Unicode characters that show up later as mysterious diffs or a search that will not match.
Text copied out of an AI assistant carries formatting that does not survive the journey. Markdown asterisks and hashes turn into literal characters in a CMS field, curly quotes and em dashes break a CSV or a code comment, and non-breaking spaces and zero-width characters ride along invisibly, showing up later as mysterious diffs or as a search that will not match.
Paste the text and pick what to remove. Each cleaner is a separate switch — strip markdown syntax, straighten quotes and dashes, normalise whitespace, remove emoji, collapse blank lines, drop invisible Unicode — and a counter shows how many characters each one changed, so you can see exactly what was in there. The zero-width pass is the one people are usually surprised by: those characters are genuinely invisible in every editor, and this is the quickest way to find out whether your text has them.
Strip markdown, straighten quotes and dashes, normalise whitespace, remove emoji, collapse blank lines or drop invisible Unicode — independently.
Those characters are genuinely invisible in every editor, and this is the quickest way to find out whether your text has them.
A counter shows how many characters each cleaner changed, so you can see exactly what was in there.
Curly quotes and em dashes that break a CSV or a code comment become their plain ASCII equivalents.
Asterisks and hashes are removed rather than left as literal characters in a CMS field.
Drop in whatever you copied out of an assistant, a document or a web page.
Switch on the passes you want. Each one lists how many characters it changed so nothing is removed silently.
Look at what was found, particularly the invisible characters, before you copy the result.
Take the cleaned output to your CMS, spreadsheet or editor.
They are Unicode characters with no visible width, such as the zero-width space and the byte order mark. They arrive through copy and paste from web pages and rich text, and they break exact-match searches and produce diffs that look identical on screen.
No, and it is not built for that. Removing markdown and normalising punctuation changes formatting, not the writing. AI detectors are unreliable in both directions anyway, and this tool is for making text portable between systems.
Because they break things downstream. A curly apostrophe in a CSV, a code comment or a legacy database field can produce mojibake or a parse error, and a straight equivalent is safer wherever the encoding is not guaranteed.
Markdown stripping unwraps code fences and inline backticks and flattens table pipes to plain text. If you want the code formatting kept, leave the markdown pass switched off and use the other cleaners.
What people who use the AI Text Cleaner think of it. Reviews come from visitors and are posted as written.
No ratings yet
Yours would be the first.
0 reviews
Newest first
No reviews yet
Be the first to rate the AI Text Cleaner. It takes half a minute and no account.
Write the first reviewBuild a Stable Diffusion negative prompt from the problems you keep getting.
Score a prompt on the parts that change the answer, and see what is missing.
Turn question and answer pairs into a JSONL training file for OpenAI, Llama or Gemini.
See the file size of every GGUF quant of a model, and which ones fit your memory.
Estimate how many tokens your text costs before you send it.
Split long documents into overlapping chunks for embeddings and RAG.
Write both custom instruction boxes, each checked against the 1,500 character limit.
Compare two embedding vectors by cosine similarity, dot product and distance.