humanizer
Skill8 starsChanged 49 days agoMITView on GitHub
Removes AI-writing tells from any draft: em dashes, buzzwords and corporate filler, throat-clearing transitions, manufactured candor, forced rule-of-three, "it's not X, it's Y" constructions, reveal-bait setups, and other statistical-average patterns. Use whenever writing or polishing prose, and always as the final pass on anything written in a specific voice. Also invoke directly on any draft ("run the humanizer on this," "de-AI this"), including text a human wrote or pasted from another tool. Voice-neutral: it strips tells without imposing a specific voice.
188 other projects use a version of this.
Adopted by
Most recently taken up by
- aditya-vermax/skillstheir own version, first seen Nov 8, 2026
- chrisleekr/skillstheir own version, first seen Nov 8, 2026
- xhqing/CommunityManagerAgenttheir own version, first seen Nov 8, 2026
- CloudChef/atlasclaw-providerstheir own version, first seen Nov 8, 2026
- Fractera/Agent-Engineering-Infrastructuretheir own version, first seen Nov 8, 2026
- Fractera/Agentic-Engineering-Infrastructuretheir own version, first seen Nov 8, 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 4 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 versions188 projects use a version of this.
Word for word · 0
None found.
Edited · 0
None found.
Wrote their own version · 188
- NousResearch/hermes-agent★ 262ksame name
- blader/humanizer★ 63ksame name
- QuivrHQ/quivr★ 40ksame name
- The-Vibe-Company/quivr★ 40ksame name
- rommapp/romm★ 14ksame name
- github/gh-aw★ 5.6ksame name
- latitude-dev/latitude-llm★ 4.7ksame name
- github/squad★ 3.4ksame name
- compozy/compozy★ 2.9ksame name
- SRombauts/SQLiteCpp★ 2.8ksame name
178 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 repository4 other files it gives its agents.
- assessSkill
- code-watchSkill
- pre-commitSkill
- systematic-debuggingSkill
The file
We split this skill into the rules it gives an agent
3 rules. 0 are common, 3 are distinctive, and 0 are very specific to this project.
0
Common
Most of the 175 edited versions we read keep these.
3
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 68
Don't manufacture a reveal.
- Distinctive: line 77
Don't use "load-bearing" to mean important or central.
- Distinctive: line 81
Don't pad a sentence to three parallel items when two would do, or invent a third for rhythm.
“Common” is measured: kept by at least 6 in 10 of the 175 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
The single source of truth for voice-neutral AI-writing tells. These are patterns a language model produces by default because it is averaging its training data. They have nothing to do with any one person's voice.
Why this is its own skill. Most people end up with more than one voice guide: a personal one, a company one, maybe a docs one. Each will accumulate its own copy of the same anti-pattern list, and those copies drift. Keep the shared tells here, keep only genuinely voice-specific rules in each voice guide, and have each voice guide point at this file as its final pass. Same single-source rule you'd apply to code.
How to use it: scan the draft against every category below, rewrite the hits, and when asked, show what changed and why. The point is catching them all in one pass. A person can hold about five of these in mind; this list holds all of them.
When you find a new tell worth banning permanently, add it here rather than to a voice guide. Add only tells that show up in real drafts. A bloated checklist makes writing read like it was built to satisfy a rubric, which is its own tell.
1. Em dashes: do not use
A hard rule. Em dashes (—) are the single most recognizable AI-writing tell.
Replace them with:
- A period. Break the thought into two shorter sentences. Usually the best option.
- A comma for a lighter pause or a parenthetical clause.
- A colon to introduce an explanation or list.
- Parentheses for an aside.
- Restructure the sentence entirely if none of the above work cleanly.
Examples:
- BAD: "The migration — the one we postponed twice — is now blocking the release."
- GOOD: "The migration, the one we postponed twice, is now blocking the release."
- ALSO GOOD: "The migration is now blocking the release. We postponed it twice."
2. Buzzwords and corporate filler
Cut on sight, and replace with the plain word or the concrete thing:
- "synergy" / "synergies"
- "leverage" (use "use") and "leverage our learnings"
- "move the needle"
- "best-in-class"
- "robust"
- "holistic"
- "value-add" (unless quoted, deliberately)
- "innovative" (show it, don't say it)
- "transformative" (show it, don't say it)
- "In today's rapidly evolving landscape..."
- "It is worth noting that..."
- "It is important to recognize..."
3. Throat-clearing and filler transitions
Phrases that announce the next sentence instead of delivering it. Cut them and start the actual sentence:
- "Here's the thing" / "Here's the thing though"
- "Let me explain." / "Let me break this down."
- "Here's the deal." / "Here's what I mean."
- "The truth is..."
- "Make no mistake."
- "Let's dive in." / "Buckle up."
4. Manufactured candor
Don't use "honest," "honestly," "the honest truth," "let's be honest," "to be honest," or "the honest reason" to add weight or signal candor. It undercuts authority rather than building it. The word is fine in its literal sense (an honest answer, an honest assessment). The tic is the flourish.
Same ban on reveal-style headings that manufacture candor: "The honest reason to do X," "The real truth about Y," "The thing nobody tells you." Use plain descriptive headings instead ("Why now," "What this costs").
5. Reveal-bait and grandiose setups
Don't manufacture a reveal. State the point plainly and let it carry itself. Ban:
- "the one thing everyone misses"
- "the detail that flips the whole argument"
- "here's the part the headline buries"
- "now the turn"
- "the thing worth saying out loud," "worth saying out loud," or any "[X] out loud" construction. Absolute ban. To make a point emphatically, make it directly, or use a plain heading like "What it comes down to."
6. "load-bearing" as a metaphor
Don't use "load-bearing" to mean important or central. Replacements: "central," "the deciding factor," "critical," "the thing that matters most," or restructure to make the point directly.
7. Forced rule-of-three
Don't pad a sentence to three parallel items when two would do, or invent a third for rhythm. Three is fine when there genuinely are three things. It's a tell when the third is filler ("faster, cheaper, and better").
8. "It's not X, it's Y"
The "it's not about X, it's about Y" / "this isn't X, it's Y" construction is a heavy tell. Rewrite to state Y directly, or contrast without the formula.
9. Reflexive hedging
Qualifiers stacked out of habit ("may," "might," "could potentially," "it could be argued") drain authority. Cut hedging that exists only as a verbal tic.
Where genuine uncertainty or optionality exists, a hedge is correct and cutting it makes the writing dishonest. How much hedging a piece should carry is voice-specific, so leave that calibration to your voice guide. This skill only flags hedging-as-tic.
10. Closing filler
No "In conclusion," "To sum up," or "To summarize." Just end. The last line should land, not announce itself.
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.

