The GPTCleanupTool blog archive

Every article published, in one plain list.

The complete archive of GPTCleanupTool articles: every guide on hidden characters, invisible Unicode and cleaning AI text, listed in one place.

Purpose

What the blog archive is

The blog page introduces what this site writes about and gives each article some context. This page skips the context and lists every article that exists, in full, so it stays useful even after the blog itself has moved on to introducing something newer.

Nothing here is trimmed or teased. If an article is listed below, it is published and complete.

The complete list

Every article published so far

Two articles exist right now, both on invisible characters in AI-written text.

How to remove hidden characters from ChatGPT text
Goes through the full list of invisible characters the site's cleaner checks for, why chat interfaces leave them behind, what each one breaks once pasted elsewhere, from a broken URL slug to a JSON file that will not parse, and the three steps to remove them without touching a single visible word.
Zero-width characters and why they get flagged
Narrows in on the zero-width family specifically: what they are, why scanners flag them, and the honest caveat that some of them, the joiner inside an emoji and the non-joiner in Persian and Indic scripts, are doing real work rather than sitting there as residue. It ends with a way to clean the residue without stripping the characters that are actually needed.

Upkeep

How an article gets added to this archive

An article is added here once it is finished, checked against the code it describes, and published on the blog. There is no separate draft or preview state visible on the site, so if it is not on this page, it does not exist yet.

That check against the code is not a formality. Before either article above was published, someone opened the actual list of characters the cleaner scans for and counted it rather than estimating, because a wrong count in an article about precision would undercut the article's own point.

Accuracy

Why an old article gets rewritten rather than replaced

If the tool an article describes changes, the article gets rewritten to match rather than left as it was or quietly deleted. The list of characters the cleaner scans for is the clearest example: if that list changes, the article describing it is wrong until it is corrected, regardless of how well it still reads.

This matters more here than on an ordinary blog, because these articles cite specific facts about a specific piece of code rather than general opinion. An opinion can age; a wrong character count just becomes an error, and errors get fixed rather than left standing because the surrounding sentences are otherwise fine.

Scope

What this archive does not include yet

There are no articles here on image or video watermarks, which our media workflow covers on its own pages. There is nothing on pricing, because none of the tools on this site have any. And there is nothing framed as a guaranteed way past a specific AI detector, because no article makes that claim.

There is also no general writing advice, no roundup of other tools, and no comparison chart. Keeping the archive narrow to what this site actually does is a deliberate choice rather than a gap waiting to be filled with unrelated content.

How to use it

How to use the archive alongside the tools

Read the entry that matches your situation, open the full article if you want the reasoning, then go to the actual tool, the homepage's watermark remover and checker, the accent remover, or the general ChatGPT text cleaner, to clean the text itself. The archive explains; the tools act.

If you already know what a zero-width space is and just need it gone, you do not need to read anything on this page first. The archive exists for the moment before that, when something in a pasted block of text is behaving strangely and you do not yet know why.

How this differs

How the archive differs from the blog page

The blog page is written to introduce the subject and give each article a reason to open it. This page is written to be complete and stable: a single reference list you can return to once you already know what you are looking for.

Bookmark this page rather than the blog page if what you want is a durable link to everything published, since the archive's job is to stay the same shape even as its own contents grow.

FAQ

Questions about the archive

Everything below is answered in full, nothing to click open.

Is this every article on the site?

Yes. If an article exists and is published, it is listed on this page.

How is this different from the main blog page?

The blog page introduces the subject and gives context for each article. This page is the plain, complete list without the surrounding introduction.

How many articles are there?

Two, both listed above with a short description of what each one covers.

Are old articles ever removed from this archive?

Not typically. If a tool changes, the article describing it is rewritten to stay accurate rather than deleted.

Is the archive free to read?

Yes. There is no signup or paywall anywhere on this site.

Does the archive cover image or video watermark guides?

No. Image and video work lives in our media workflow. This archive is text only.

Do the archived articles apply to models other than ChatGPT?

Yes. Both cover characters and behaviour that show up in text from any chat interface, not one model specifically.

Can I read an article without using the tools?

Yes. Reading is free and does not require running anything.

Does reading an archived article change my text?

No. The articles are reference material. Cleaning happens on the actual tool pages.

Will an article here promise a specific AI detector score?

No. Neither article makes that promise, and none added later will either.

Is there a way to sort or filter the archive?

With two articles, sorting is not needed yet. Both are listed in full above.

Is there an RSS feed for new articles?

Not yet.

How do I know an archived article is still accurate?

It is rewritten whenever the tool it describes changes, so a listed article is treated as current unless the site says otherwise.

Can I suggest a topic for a future article?

Yes, through the contact page.

Does the archive work on mobile?

Yes. The list and each linked article read the same way on a phone.

Do I need an account to browse the archive?

No. There is no account system on this site.

Why does the archive list full descriptions instead of just titles and dates?

With only two articles, a full description is more useful than a bare title, and it lets you decide whether an article is relevant without opening it first.

What should I read first if I have never dealt with a hidden character before?

Start with the guide to removing hidden characters from ChatGPT text. It covers the full list before the second article narrows in on one family of it.

Two articles, listed in full.

Open the one that matches your situation, then clean the text on the actual tool page.

Open the tools