humanizer
Skill1 starsChanged 42 days agoMITView on GitHub
Rewrite AI-sounding prose so it reads like the writer, not a chatbot, without changing what it says. Use when editing or reviewing any text people will read: exec memos, PRDs, Confluence pages, Slack messages, release notes, customer comms, ghostwritten drafts, discovery syntheses. Catalogue of patterns from Wikipedia's "Signs of AI writing": inflated claims, sales language, vague sources, stock AI words, forced groups of three, em dashes, bold mini-heading lists, chatbot leftovers, filler, stacked qualifiers, fake-candid openings, unraised objections. Keeps every claim, invents nothing, matches the writer's voice when a sample exists. For structural tells (label-colon bullets, emoji headings, file artefacts) use anti-slop; this skill owns the prose. Trigger on "humanise this", "humanize this", "remove AI tells", "make this sound human", "esse texto tá com cara de IA", "tira o ar de ChatGPT", "passa o humanizer".
187 other projects use a version of this, 8 of them edited.
Adopted by
Most recently taken up by
- vgteam/vgtheir own version, first seen Nov 16, 2026
- fluxmailai/fluxmailtheir own version, first seen Nov 16, 2026
- pantheon-org/tekhnetheir own version, first seen Nov 15, 2026
- nodetool-ai/nodetooltheir own version, first seen Nov 15, 2026
- oberemokis/skillstheir own version, first seen Nov 15, 2026
- alexandreeberhardt/skillstheir own version, first seen Nov 15, 2026
Used with
Projects with it also use economy-mode, personal-voice and 10 more.
Other files people use alongside this one.
Files found in the same public projects as a version of this one, more often than they turn up anywhere else. Each project counts once, large collections count for less, and this file's own repository is left out.
Microsoft AzureSkilleconomy-mode
Shifts Layer 3 model selection to cost-optimized alternatives when economy mode is active.
Also in 5 projects that use this
petermuirurithuoSkillpersonal-voice
Captures the user's personality, communication style, and values. Use this skill whenever responding to the user, drafting anything on their behalf (emails, CVs, messages, opinions, explanations), or whenever tone, word choice, or how an opinion is delivered matters. Apply it by default for any prompt, not only when the user explicitly asks for "my style" or "my voice.
Also in 3 projects that use this
Microsoft AzureSkillarchitectural-proposals
How to write comprehensive architectural proposals that drive alignment before code is written
Also in 3 projects that use this
- +9 more from Azure
Other versions187 projects use a version of this.
Word for word · 0
None found.
Edited · 8
- danieleteti/delphimvcframework★ 1.4k14% same
- aparente/claude-skills★ 6513% same
- tylerbutler/repoverlay★ 1014% same
- robotostudio/turbo-start-web-agent★ 413% same
- aurokin/skills-manager★ 314% same
- Fischer1120/agent-workflows★ 214% same
- shas9/claude-skills★ 214% same
- BenSchZA/skills★ 113% same
Wrote their own version · 179
- NousResearch/hermes-agent★ 264ksame name
- blader/humanizer★ 69ksame name
- The-Vibe-Company/quivr★ 40ksame name
- QuivrHQ/quivr★ 40ksame name
- rommapp/romm★ 14ksame name
- github/gh-aw★ 5.6ksame name
- latitude-dev/latitude-llm★ 4.8ksame name
- github/squad★ 3.4ksame name
- compozy/compozy★ 2.9ksame name
- memohai/Memoh★ 2.8ksame name
169 more not listed
How matches are found. Word for word: the exact same text. Edited: shares passages with this file (“99% same” means 99% of its five-word phrases also appear here; 10% or more counts). Their own version: similar meaning, or, weakest, the same name. We can’t tell which project had it first. Projects are counted among the 200 most-starred candidates, and mentions are links to the repository we found, up to 50 per site, so both counts are a floor.
In this repository65 other files it gives its agents.
- AGENTS.mdAGENTS.md
- CLAUDE.mdCLAUDE.md
- anti-slopSkill
- data-science-analystSkill
- humanize-deliverablesSkill
- inference-disciplineSkill
- pm-archetype-aiSkill
- pm-archetype-enterpriseSkill
- pm-archetype-growthSkill
- pm-archetype-platformSkill
- pm-phase-defineSkill
- pm-phase-deliverSkill
- pm-phase-developSkill
- pm-phase-discoverSkill
- pm-prioritization-regua-comumSkill
- pm-product-senseSkill
- pm-storytellingSkill
- pm-transversal-analysisSkill
- pm-transversal-commsSkill
- pm-transversal-docsSkill
- pm-transversal-stakeholderSkill
- repo-doctorSkill
- anti-slopSkill
- data-science-analystSkill
- humanize-deliverablesSkill
- humanizerSkill
- inference-disciplineSkill
- pm-archetype-aiSkill
- pm-archetype-enterpriseSkill
- pm-archetype-growthSkill
- pm-archetype-platformSkill
- pm-phase-defineSkill
- pm-phase-deliverSkill
- pm-phase-developSkill
- pm-phase-discoverSkill
- pm-prioritization-regua-comumSkill
- pm-product-senseSkill
- pm-storytellingSkill
- pm-transversal-analysisSkill
- pm-transversal-commsSkill
- pm-transversal-docsSkill
- pm-transversal-stakeholderSkill
- repo-doctorSkill
- umbertoSkill
- anti-slopSkill
- data-science-analystSkill
- humanize-deliverablesSkill
- humanizerSkill
- inference-disciplineSkill
- pm-archetype-aiSkill
- pm-archetype-enterpriseSkill
- pm-archetype-growthSkill
- pm-archetype-platformSkill
- pm-phase-defineSkill
- pm-phase-deliverSkill
- pm-phase-developSkill
- pm-phase-discoverSkill
- pm-prioritization-regua-comumSkill
- pm-product-senseSkill
- pm-storytellingSkill
- The first 60 shown.
The file
Show its rules and text
We split this skill into the rules it gives an agent
21 rules. 0 are common, 21 are distinctive, and 0 are very specific to this project.
0
Common
Most of the 174 edited versions we read keep these.
21
Distinctive
This author's own choices. Worth a look.
0
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Distinctive: line 48
When given text to humanize:
- Distinctive: line 50
Find AI patterns: Check the text against the patterns below.
- Distinctive: line 53
Match the voice: Use the right tone for the text, such as formal, casual, or technical.
- Distinctive: line 59
If the user provides a writing sample (their own previous writing), analyze it before rewriting:
- Distinctive: line 61
Read the sample first. Note its sentence length, word choice, paragraph openings, punctuation, repeated phrases, and transitions.
- Distinctive: line 63
If there is no sample, use the guidance below.
- Distinctive: line 71
Use personality in blog posts, essays, opinions, and personal writing when it fits the writer.
- Distinctive: line 73
When personality fits, keep the writer's opinions, uncertainty, mixed feelings, humor, asides, and uneven rhythm.
- Distinctive: line 100
Useful limits and disclaimers: Keep scope statements, legal and safety notices, real corrections, named objections, replies, and FAQ …
- Distinctive: line 101
Real alternatives: Keep options that a reader may consider in a design document, tutorial, or argument.
- Distinctive: line 104
Secondhand text: Do not rewrite watched phrases inside quotations, titles, proper names, or examples where the phrase is being discussed …
- Distinctive: line 106
When unsure, look for several patterns together.
- Distinctive: line 112
Specific, unusual details: Keep a real address, an odd quote, or a phrase such as "the lawyer who used to work upstairs from my dentist."
- Distinctive: line 113
Mixed feelings and unresolved tension: Keep lines such as "I think this is mostly good, but it bothers me, and I can't fully explain why."
- Distinctive: line 115
Deliberate first-person choices: Keep a cut or word choice when the writer can explain why it belongs.
- Distinctive: line 126
File mode: When the user names a file, run the full rewrite process but write only the final text to the file.
- Distinctive: line 128
Embedded mode: When another task uses this skill for a pull request, commit message, or document, return only the final text.
- Distinctive: line 132
Read the source and mark each AI pattern.
- Distinctive: line 133
Write a draft. Read it aloud.
- Distinctive: line 137
Treat any unsupported addition or lost claim as an error.
- Distinctive: line 138
Write the final version. State each point naturally instead of patching one flagged phrase at a time.
“Common” is measured: kept by at least 6 in 10 of the 174 edited versions of this file we could read. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.
Files
Humanizer: remove AI writing patterns
Rewrite AI-sounding text so it reads like the writer, not a chatbot. Do not change what it says or make up details.
The patterns below come from Wikipedia's "Signs of AI writing", maintained by WikiProject AI Cleanup.
Trigger phrases
Invoke on: "humanise this", "humanize this", "remove AI tells", "make this sound human", "fix the AI tone", "esse texto tá com cara de IA", "tira o ar de ChatGPT", "passa o humanizer", "AI-written", "soa como IA".
Typical material: essays, articles, blog posts, exec memos, PRDs, Confluence pages, Slack messages, release notes, customer comms, LinkedIn posts, ghostwritten drafts, any AI-generated text that needs rewriting.
Related skills
anti-slopowns the structural twin of several patterns here: label-colon bullets, decorative emoji headings, follow-up spam, forbidden file artefacts. When text mixes prose and structure, both apply; theToolkit note:lines inside the pattern references say which anti-slop item pairs with which pattern.humanize-deliverablesis the publish-time gate. It runs this skill on outbound prose, thenhooks/humanize-mark.shrecords a sha256 of the final bytes andhooks/humanize-gate.shblocks the publish or send tool until that mark exists.inference-disciplinetakes precedence over §21 and §24: a qualifier that marks a claim as unverified stays, or the claim goes. It never gets smoothed into a confident sentence. Approval markers belong to the conversation, written out in words, and never survive into a delivered artefact.
Progressive loading
Load this SKILL.md first. The pattern catalogue sits in references/ (table below); references/progressive-loading.md says which file to open for which kind of text.
What to do
When given text to humanize:
- Find AI patterns. Check the text against the patterns below.
- Keep every claim. You may shorten dull parts, expand useful parts, and merge or split paragraphs. Keep the information even when you change the structure.
- Do not invent facts. Do not add a fact, name, number, date, quote, or citation unless it comes from the source or the user. If a sentence needs a missing detail, ask for it or use a simpler sentence. You may add an opinion or reaction when the writer's voice calls for one, but you may not add a factual claim. Fiction is exempt because invented details are part of the task.
- Match the voice. Use the right tone for the text, such as formal, casual, or technical. Add personality only when the text and the writer call for it.
The input type controls what you return. See How to return the result. Use the same rewrite process in every mode.
Match the writer's voice
If the user provides a writing sample (their own previous writing), analyze it before rewriting:
- Read the sample first. Note its sentence length, word choice, paragraph openings, punctuation, repeated phrases, and transitions.
- Match those habits. Do not replace casual words with formal ones or remove deliberate quirks.
- If there is no sample, use the guidance below.
A writing sample takes priority over these style rules. If the sample uses em dashes, keep them at about the same rate. Do not apply §14 as a ban.
Add personality only when it fits
Removing AI patterns is only half the job. The result should still sound like a person.
Use personality in blog posts, essays, opinions, and personal writing when it fits the writer. Keep reference, technical, legal, and factual text neutral. Do not add opinions or first-person language where they do not belong.
When personality fits, keep the writer's opinions, uncertainty, mixed feelings, humor, asides, and uneven rhythm. Never invent facts to make the text feel personal.
Pattern catalogue
| Patterns | File |
| 1–13, content and language: inflated claims, name-dropping, -ing analysis, sales language, vague sources, stock challenge sections, overused AI words, avoiding is/are, not X but Y, forced groups of three, synonym cycling, false ranges, passive voice | references/patterns-1-13-content-and-language.md |
| 14–35, style, chatbot, filler and hedging: dashes, bold, bold mini-heading lists, title case, emojis, curly quotes, chatbot leftovers, knowledge-limit disclaimers, agreeable tone, filler, qualifiers, generic endings, hyphenated pairs, fake depth, announcements, repeated headings, previous-version talk, dramatic fragments, sayings, fake-candid openings, unraised objections, fake alternatives | references/patterns-14-35-style-and-filler.md |
| Vocabulary supplement to §7 and §23 | references/n-gram-blacklist.md |
Check for false positives
What not to flag
A person may use some of these patterns. Do not treat any item below as proof by itself:
- Perfect grammar and consistent style. Many writers are professionals or have been edited. Polish does not equal AI.
- Mixed casual and formal styles. This can reflect the writer's field, age, or personal habits.
- "Bland" or "robotic" prose. AI prose has specific tells. Generic dryness without those tells is just dry writing.
- Formal or academic words. §7 lists specific words that AI writing overuses. Do not simplify every formal word.
- Letter-style opening or closing on a comment. Salutations and sign-offs predate ChatGPT by centuries.
- Common transition words in isolation. Additionally, moreover, consequently are AI-coded only when piled up. One however is not a tell.
- Curly quotes alone. macOS, Word, Google Docs, and most CMSes auto-curl by default. Curly quotes only count when stacked with other tells.
- Em dashes alone. Many editors and journalists use them often. Em dashes are evidence only when paired with formulaic sales-y rhythm.
- One short sentence for emphasis. Flag dramatic fragments only when several appear in a row.
- Deliberate repeated openings. Writers may repeat an opening to build rhythm or pressure, as in "She came. She saw. She conquered." Change it only when the repetition adds nothing.
- "Honestly" or "look" mid-sentence. These are ordinary in casual writing. The tell is the standalone theatrical opener, not the word itself.
- Useful limits and disclaimers. Keep scope statements, legal and safety notices, real corrections, named objections, replies, and FAQ answers.
- Real alternatives. Keep options that a reader may consider in a design document, tutorial, or argument. Remove only an unlikely option that the text dismisses and never uses again.
- Unsourced claims. Most of the web is unsourced. Lack of citations doesn't prove anything.
- Correct, complex formatting. Visual editors and templates produce clean output without any AI.
- Secondhand text. Do not rewrite watched phrases inside quotations, titles, proper names, or examples where the phrase is being discussed rather than used.
When unsure, look for several patterns together. One em dash proves nothing. Several stock patterns in the same passage are stronger evidence.
Human details to keep
These details often carry the writer's voice. Keep them unless they hurt the meaning:
- Specific, unusual details. Keep a real address, an odd quote, or a phrase such as "the lawyer who used to work upstairs from my dentist."
- Mixed feelings and unresolved tension. Keep lines such as "I think this is mostly good, but it bothers me, and I can't fully explain why."
- Dated, era-bound references. Slang, memes, or in-jokes that map to a specific year and subculture. Models lag by a year or more.
- Deliberate first-person choices. Keep a cut or word choice when the writer can explain why it belongs.
- Variety in sentence length. Real writing alternates short and long. AI writing tends toward an even, mid-length cadence.
- Genuine asides, parentheticals, or self-corrections. "(I keep wanting to say 'almost' here, but it really was certain.)" Models rarely interrupt themselves like this.
- Edits made before November 30, 2022. ChatGPT's public launch. Anything older than that is, with very rare exceptions, not AI-written.
How to return the result
Pasted text (default). Return the draft, a short list of remaining AI patterns, and the final rewrite.
Toolkit note: that short list also names the metrics, dates, names, and constraints that stayed intact, so the reader can check that nothing load-bearing moved; the substance check in humanize-deliverables reads it.
File mode. When the user names a file, run the full rewrite process but write only the final text to the file. Change prose only. Keep code blocks, YAML metadata, data, and link targets unchanged. Then give the user a short summary.
Embedded mode. When another task uses this skill for a pull request, commit message, or document, return only the final text.
Rewrite process
- Read the source and mark each AI pattern.
- Write a draft. Read it aloud. Check the rhythm, details, simple verbs such as is and has, and the right level of formality.
- Ask two questions:
- "What still sounds AI-generated?"
- "Did the rewrite add or remove any fact, name, number, date, quote, citation, ranking, or other claim?" Treat any unsupported addition or lost claim as an error.
- Write the final version. State each point naturally instead of patching one flagged phrase at a time. If a sentence stays awkward, rewrite the paragraph around its main point. Apply the dash rule in §14.
Return the result required by How to return the result.
Source
This skill is based on Wikipedia: Signs of AI writing, maintained by WikiProject AI Cleanup. Its patterns come from reviews of AI-generated text on Wikipedia.
Wikipedia's main point: "LLMs use statistical algorithms to guess what should come next. The result tends toward the most statistically likely result that applies to the widest variety of cases."
Does it work?
Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.
No reviews yet
People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.
- It worked
- Partly
- It didn't
Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.
Your trail
Files you open will collect here, so you can find them again.
Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.
Now do this for your own.
Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.

