humanizer
Skill27 starsChanged 8 months agoApache-2.0View on GitHub
Detect and rewrite AI writing patterns to make content sound authentically human. Supports voice calibration from the author's own writing samples. Use when asked to humanize text, remove AI patterns, calibrate voice, audit content for AI detectability, or check if writing sounds like ChatGPT. Triggers: /humanize, /voice-calibrate, /audit-ai
187 other projects use a version of this.
Adopted by
Most recently taken up by
- 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
- mad01/thismoontheir own version, first seen Nov 15, 2026
- nuwrldnf8r/book_editor_skilltheir 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 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 versions187 projects use a version of this.
Word for word · 0
None found.
Edited · 0
None found.
Wrote their own version · 187
- NousResearch/hermes-agent★ 264ksame name
- blader/humanizer★ 69ksame 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
- felinics/Memoh★ 2.8ksame name
177 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 repository40 other files it gives its agents.
- AGENTS.mdAGENTS.md
- CLAUDE.mdCLAUDE.md
- webapp-testingSkill
- env-hubSkill
- mcp-builderSkill
- mintlify-docs-updaterSkill
- reviewSkill
- shipSkill
- skill-creatorSkill
- specSkill
- internal-commsSkill
- ml-failure-auditSkill
- changelog-generatorSkill
- algorithmic-artSkill
- canvas-designSkill
- instagram-reel-editorSkill
- remotionSkill
- slack-gif-creatorSkill
- ab-test-setupSkill
- copywritingSkill
- instagram-postingSkill
- marketing-ideasSkill
- seo-auditSkill
- claude-for-legalSkill
- contract-review-risk-scoringSkill
- doc-coauthoringSkill
- docxSkill
- legal-work-pluginSkill
- pdfSkill
- pptxSkill
- xlsxSkill
- pm-skillsSkill
- resume-tailoringSkill
- last30daysSkill
- brand-guidelinesSkill
- excalidrawSkill
- frontend-designSkill
- open-designSkill
- theme-factorySkill
- web-artifacts-builderSkill
The file
We split this skill into the rules it gives an agent
2 rules. 0 are common, 2 are distinctive, and 0 are very specific to this project.
0
Common
Most of the 174 edited versions we read keep these.
2
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 29
Run a second-pass audit: read the rewrite cold and ask "does any sentence still sound AI-generated?"
- Distinctive: line 50
Document the fingerprint as a short style profile.
“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
Overview
Detects and rewrites 29 AI writing patterns across content, language, style, and communication categories. Patterns include: significance inflation, vague attributions, copula avoidance, synonym cycling, em dash overuse, sycophantic tone, chatbot artifacts, and more. Also supports voice calibration from your own writing samples.
Source Repository
- GitHub: blader/humanizer
- Install upstream:
npx skills add blader/humanizer
/humanize
Full 29-pattern detection and rewrite. Analyzes the input for all AI-sounding patterns, rewrites the content, then runs a second-pass "obviously AI?" audit before returning the final version.
Workflow
- Receive the text to humanize and any brand voice notes.
- Scan for all 29 patterns — flag each instance found.
- Rewrite the text: fix flagged patterns, preserve the core argument and facts.
- Run a second-pass audit: read the rewrite cold and ask "does any sentence still sound AI-generated?"
- Return the final rewritten text with a brief summary of the main changes made.
Example prompts
| Use case | Task prompt |
| Blog post | Humanize this blog post draft. Remove any AI-sounding patterns. Preserve the core argument but make it read like a person who actually has opinions wrote it. |
| Rewrite this LinkedIn post. It currently sounds like ChatGPT wrote it. Cut the significance inflation, remove the em dashes, and make it direct. | |
| Press release | Run the 29-pattern check on this press release. Flag every AI pattern you find, then rewrite it. The brand voice is confident and plain-spoken, not corporate. |
/voice-calibrate
Accepts 2–3 writing samples from the target author, extracts their stylistic fingerprint (sentence rhythm, vocabulary, punctuation habits), then applies that fingerprint to any AI-generated text.
Workflow
- Receive 2–3 writing samples from the target author.
- Analyze samples for: average sentence length, punctuation habits, vocabulary range, tonal register, structural patterns (how they open/close paragraphs), and idioms they favor.
- Document the fingerprint as a short style profile.
- Apply the fingerprint to the target AI text, rewriting to match the author's natural voice.
- Return the rewritten text with the style profile so it can be reused in follow-up requests.
Example prompts
| Use case | Task prompt |
| Newsletter | Here are 3 of my past newsletters [attached]. Use them to learn my voice, then rewrite this AI-drafted issue to match how I actually write. |
| CEO post | Calibrate to the CEO's voice using these 5 LinkedIn posts. Then rewrite this product announcement so it sounds like her, not our content team. |
| Personal brand | I want to post consistently on LinkedIn but don't have time to write from scratch. Learn my voice from these samples and rewrite these 5 AI drafts to match it. |
/audit-ai
Scores text 0–100 for AI detectability across all 29 pattern categories. Highlights specific phrases most likely to trigger AI detectors and returns a prioritized fix list ranked by severity.
Workflow
- Receive the text to audit.
- Score each of the 29 patterns 0–3 (0 = not present, 3 = severe).
- Compute an overall AI detectability score (0–100, higher = more detectable).
- Highlight the top phrases that are most suspicious.
- Return the score, highlighted phrases, and a prioritized fix list (must-fix vs. nice-to-fix).
Example prompts
| Use case | Task prompt |
| Pre-publish | Audit this article before we publish it. Give me an AI score, highlight the top 10 most suspicious phrases, and tell me which ones I absolutely must fix. |
| SEO content | Score these 5 blog posts for AI detectability. Rank them worst to best and give me a fix list for the 3 worst offenders. |
| Team workflow | Build me a pre-publish checklist based on the 29 patterns so our editors know what to review every time before anything goes live. |
The 29 AI Writing Patterns
Organized into four categories:
Content patterns: significance inflation, vague attributions, unnecessary hedging, false balance, over-explaining obvious things, unsupported universal claims
Language patterns: copula avoidance ("utilizing" instead of "using"), synonym cycling (rotating synonyms to avoid repetition), em dash overuse, passive voice stacking, gerund chains, adverb padding
Style patterns: sycophantic openers, chatbot sign-off artifacts, hollow transitional phrases ("It's worth noting that…"), conclusion telegraphing ("In conclusion…"), bullet-point everything bias
Communication patterns: fence-sitting on opinions, corporate hedging language, artificial enthusiasm, filler affirmations ("Certainly!", "Great question!"), over-structured responses, meta-commentary about the writing 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.

