| name | humanize-writing |
| description | Remove AI writing tells from publishable prose. Use for humanizing blogs, docs, release notes, tweets, email, or copy that sounds generated. |
| user_invocable | true |
| argument-hint | [text or file path to humanize] |
Humanize Writing
Detect and remove the patterns that make text recognizably AI-written. The premise (from StoryScope, arXiv:2604.03136): AI prose stays detectable even after surface cleanup, because the structural choices give it away. Fixing em-dashes and "delve" is necessary but not sufficient.
How to use
Given text (in $ARGUMENTS or a file path), run both passes below, report what you flagged, then rewrite. Don't silently rewrite, show the user which tells you found so they learn to avoid them.
Pass 1: Surface tells (fast, lexical)
These are the well-known signatures. Fix them, but know they are the easy half.
- Em-dashes and hyphens-as-dashes -- restructure with commas, semicolons, colons, or separate sentences.
- "It's not X, it's Y" contrast pattern -- banned. State Y directly.
- Overused vocabulary -- delve, tapestry, testament, realm, navigate, leverage, robust, seamless, crucial, vibrant, ever-evolving, "in today's fast-paced".
- Dead metaphors and stock phrases -- figures of speech you're used to seeing in print: "game changer", "unlock the power of", "at the end of the day", "low-hanging fruit", "double-edged sword", "the landscape of". Say it plainly or find a fresh, specific image.
- Pompous diction -- long or latinate word where a short one works: utilize (use), facilitate (help), commence (start), prior to (before), in order to (to), "a variety of" (give the number or drop it). Never a long word where a short one will do.
- Agentless passive -- "it was decided", "mistakes can be made", "improvements were introduced". Name the actor when the actor matters.
- Padding constructions -- "the fact that", "there is/are ... that", nominalizations ("perform an installation of" instead of "install"). If a word can be cut, cut it.
- Hedging filler -- "it's worth noting", "it's important to", "that said", "moreover", "furthermore" as paragraph openers.
- False-sincerity openers -- "honestly", "to be honest", "frankly", "in all honesty", "let's be real". They signal nothing and pad the sentence. Just state the point.
- Rule-of-three everywhere -- AI defaults to triples ("fast, reliable, and scalable"). Vary list length; use two or four.
- Tidy closing summary -- the "In conclusion" / "Ultimately" wrap-up that restates what was just said.
Pass 2: Structural tells (the durable ones)
This is what actually separates human from AI prose. Editing these requires real rewrites, not find-and-replace.
- Over-explains the takeaway. AI spells out the moral far more than humans do. Cut the sentence that says "the lesson here is...". Trust the reader to infer it from the example.
- Progressive disclosure. Keep information light by default. Lead with what the reader needs now. Add detail only when it helps the next question or decision. Go deep only when the text explicitly intends a deep dive.
- Vague allusions instead of specific references. Humans name real tools, versions, people, repos, numbers; AI stays generic. Replace "a popular framework" with "Nuxt 4", "studies show" with the actual source, "significantly faster" with the real figure.
- Writes as though no one is watching. Humans address the reader directly and break the fourth wall ("you've probably hit this"). AI narrates into the void. Add direct address where natural.
- Tidy, single-track structure. AI marches clue to reveal in a straight line with no loose ends. Humans digress, backtrack, leave threads open. Allow an aside or a non-linear jump.
- Narrow repertoire / over-determination. AI converges on safe defaults and resolves everything neatly. Let tension stay unresolved; admit what you don't know; allow an opinion that isn't perfectly balanced.
- Elegant variation. AI swaps synonyms for the same thing to sound sophisticated ("the tool... the utility... the solution"). Humans repeat the term. Use the same word for the same thing throughout.
- Uniform sentence rhythm. AI writes evenly weighted, medium-length sentences, one point each. Humans are bursty: a two-word punch, then a long rambling clause. Vary sentence length deliberately.
Adapt to content type
Not every tell applies everywhere. Weight by genre:
- Tweets / social: surface tells + specificity + direct address matter most. Skip structural nonlinearity.
- Release notes / docs: specificity (real version numbers, PR links) and cutting the over-explained takeaway matter most. Keep structure clear, drop the moralizing.
- Blog posts / essays: all of Pass 2 applies. This is where structure shows the most.
- Email: direct address and dropping hedging filler matter most.
Output
- List the tells you found, grouped by pass, quoting the offending phrase.
- Provide the rewritten text.
- If rewriting changed meaning anywhere, flag it explicitly rather than assuming the user's intent.
Guardrail
Break any rule above sooner than write something stilted. The goal is prose that reads human, and humans keep deliberate voice, rhythm, humor, and the occasional ornament. If a "tell" is doing real work (a metaphor that lands, a triple with punch), keep it and say why. Preserve the user's meaning, tone, and explicit constraints; never sand text into flat sameness.
