unslop · writing guide

Humanize AI text without taking our word for it

unslop scores how much a text sounds like machine writing, from 0 to 100, and then rewrites it the way a person would say it. The rewrite comes back with the score before and the score after, so you can see the difference instead of taking our word for it.

Open the demo → See the pricing

What "humanize" means here

It means removing the habits that make a text read like a language model. It does not mean hiding the text, spinning the wording, or making a claim the reader cannot check. Concretely, unslop does three things:

  1. It finds the specific places where the text sounds mechanical, and marks them in place.
  2. It scores the text from 0 to 100 on how much of that mechanical writing it contains.
  3. It rewrites the text and returns the new score next to the old one.

It does not tell you who wrote a text, and it does not tell you whether a teacher, a client or a reviewer will accept it. It tells you what is wrong with the writing itself, which is the part you can fix.

What it looks for

Every rule below is one the engine checks for on every run, in English and in Spanish.

Phrases that carry no information. "It's worth noting", "in today's fast-paced world", "in conclusion", "dive into", "a wide range of", "cabe destacar", "es importante mencionar", "en conclusión". These are the most expensive hits per word because a phrase built to be generic almost never appears in human writing by accident.

Vocabulary that belongs in a press release. "delve", "leverage", "seamless", "robust", "landscape", "holistic", "synergy", "paradigm", "myriad", "panorama", "sinergia", "holístico". A single word can be legitimate, which is why words score less than phrases.

Structure that gives it away. "Not just X, but Y". "It's not about X, it's about Y". "More than X, it's Y". A rule of three where three adjectives of the same length sit side by side ("clear, direct and honest"). A chain of em dashes used as a dramatic pause. A rhetorical question followed by a one-word answer. Hedging that refuses to commit ("arguably", "to some extent", "it depends"). The assistant sign-off ("I hope this helps", "let me know if").

Shape. The habits a model gives a document without noticing: every section opens with the same kind of sentence, every bullet starts with a bolded label, every paragraph is the same length, and the text closes with a summary of itself.

How to use it

  1. Paste the text. Up to 5,000 characters per run on the free tier.
  2. Read the score and the marks. Each hit is highlighted where it appears, with the name of the rule that caught it, and the breakdown is split by category.
  3. Rewrite it as human. The result comes back with the score before, the score after, and the number of hits removed.

The surface pass is instant. The deep read, which is what catches the structural tells, takes up to about 40 seconds on dense text.

What the free tier includes

Paid plans raise the daily quota and the character limit, and the Team plan adds API access. The pricing page has the current numbers.

What it does not do

Questions people ask

Is it free to try? Yes. The free tier covers 10 scores and 10 rewrites a day, up to 5,000 characters, with no account and no key.

Will it make my writing sound human? It removes the patterns that read as machine writing and leaves the rest of your text alone. It is an editor, not a writer: it does not invent arguments, facts or sources.

Which languages? Spanish and English. The French phrase and vocabulary lists are included as well.

Does it work on a whole document? On any text up to 5,000 characters per run on the free tier, 40,000 on paid plans. Paste a section at a time for a long document.

Is there an API? Yes, on the Team plan only. The pricing page lists it.

Try it on your own text

Paste a paragraph you are not sure about and read the score before you read anything else. That number is the whole product.

Open the demo → See the pricing

Other pages