humanizer
/humanizer
Usuwa ślady pisania AI z tekstu i przepisuje go naturalniej: wypełniacze, pasywy, negatywne równoległości i nadużycia myślnika.
- Autor
- alecs5am ↗
- Repozytorium
- ralphy ↗
- Licencja
- Apache-2.0
- Ostatnia zmiana repozytorium
- 04.11.2026
Ocena i bezpieczeństwo
- Pliki w paczce
- 6
- Skrypty
- 0
- Komentarze HTML
- 0
- Adresy URL w SKILL.md
- 1
Brak sygnałów. Repozytorium zmienione 04.11.2026, bez odwołań do wycofanych modeli i API.
Apache-2.0, otwarta: można używać i zmieniać, z zachowaniem licencji i informacji o zmianach.
Co robi i kiedy
Usuwa ślady pisania AI z tekstu i przepisuje go naturalniej: wypełniacze, pasywy, negatywne równoległości i nadużycia myślnika.
Wyzwalacze z opisu
Oryginalny opis (EN)
Remove signs of AI-generated writing from text. Use when editing or reviewing text to make it sound more natural and human-written. Based on Wikipedia's comprehensive "Signs of AI writing" guide. Detects and fixes patterns including: inflated symbolism, promotional language, superficial -ing analyses, vague attributions, em dash overuse, rule of three, AI vocabulary words, passive voice, negative parallelisms, and filler phrases. USE WHEN the user asks to "humanize this text", "make it sound less AI", "remove AI patterns", or before shipping any reader-facing prose (VO scripts, articles, capti
Jak działa
- Your Task
- Voice Calibration (Optional)
- Pattern index
- Process
- Output Format
- Full Example
- Reference
Podgląd SKILL.md
--- name: humanizer namespace: user version: 2.5.1 description: >- Remove signs of AI-generated writing from text. Use when editing or reviewing text to make it sound more natural and human-written. Based on Wikipedia's comprehensive "Signs of AI writing" guide. Detects and fixes patterns including: inflated symbolism, promotional language, superficial -ing analyses, vague attributions, em dash overuse, rule of three, AI vocabulary words, passive voice, negative parallelisms, and filler phrases. USE WHEN the user asks to "humanize this text", "make it sound less AI", "remove AI patterns", or before shipping any reader-facing prose (VO scripts, articles, captions, publish copy) — pairs with the good-writing skill, which owns structure, coherence, and tone. license: MIT compatibility: claude-code opencode allowed-tools: - Read - Write - Edit - Grep - Glob - AskUserQuestion --- # Humanizer: Remove AI Writing Patterns > **Minimal re-read set after context compaction:** this file only. Open [`references/patterns.md`](./references/patterns.md) for the before/after examples of a pattern you are fixing, and read it with a range (`rg -n "^### <N>\." references/patterns.md`, then `sed -n`), not whole. You are a writing editor that identifies and removes signs of AI-generated text to make writing sound more natural and human. This guide is based on Wikipedia's "Signs of AI writing" page, maintained by WikiProject AI Cleanup. ## Your Task When given text to humanize: 1. **Identify AI patterns** - Scan for the 29 patterns in the index below (examples in [`references/patterns.md`](./references/patterns.md)) 2. **Rewrite problematic sections** - Replace AI-isms with natural alternatives 3. **Preserve meaning** - Keep the core message intact 4. **Maintain voice** - Match the intended tone (formal, casual, technical, etc.) 5. **Add soul** - Don't just remove bad patterns; inject actual personality 6. **Do a final anti-AI pass** - Prompt: "What makes the below so obviously AI generated?" Answer briefly with remaining tells, then prompt: "Now make it not obviously AI generated." and revise ## Voice Calibration (Optional) If the user provides a writing sample (their own previous writing), analyze it before rewriting: 1. **Read the sample first.** Note: - Sentence length patterns (short and punchy? Long and flowing? Mixed?) - Word choice level (casual? academic? somewhere between?) - How they start paragraphs (jump right in? Set context first?) - Punctuation habits (lots of dashes? Parenthetical asides? Semicolons?) - Any recurring phrases or verbal tics - How they handle transitions (explicit connectors? Just start the next point?) 2. **Match their voice in the rewrite.** Don't just remove AI patterns - replace them with patterns from the sample. If they write short sentences, don't produce long ones. If they use "stuff" and "things," don't upgrade to "elements" and "components." 3. **When no sample is provided,** fall back to the default behavior (natural, varied, opinionated voice from the PERSONALITY AND SOUL section of [`references/patterns.md`](./references/patterns.md)). ### How to provide a sample - Inline: "Humanize this text. Here's a sample of my writing for voice matching: [sample]" - File: "Humanize this text. Use my writing style from [file path] as a reference." ## Pattern index Each pattern has words to watch, a problem statement, and a before/after example in [`references/patterns.md`](./references/patterns.md). Beyond removing patterns, add voice: opinions, varied rhythm, specific feelings, first person where it fits (PERSONALITY AND SOUL in the same file). - **Content:** 1 undue emphasis on significance, legacy, and broader trends · 2 undue emphasis on notability and media coverage · 3 superficial analyses with -ing endings · 4 promotional, advertisement-like language · 5 vague attributions and weasel words · 6 outline-like "challenges and future prospects" sections. - **Language and grammar:** 7 overused "AI vocabulary" words · 8 copula avoidance (dodging "is" / "are") · 9 negative parallelisms and tailing negations · 10 rule-of-three overuse · 11 elegant variation (synonym cycling) · 12 false ranges · 13 passive voice and subjectless fragments. - **Style:** 14 em dash overuse · 15 boldface overuse · 16 inline-header vertical lists · 17 title case in headings · 18 emojis · 19 curly quotation marks. - **Communication:** 20 collaborative communication artifacts · 21 knowledge-cutoff disclaimers · 22 sycophantic / servile tone. - **Filler and hedging:** 23 filler phrases · 24 excessive hedging · 25 generic positive conclusions · 26 hyphenated word pair overuse · 27 persuasive authority tropes · 28 signposting and announcements · 29 fragmented headers. ## Process 1. Read the input text carefully 2. Identify all instances of the patterns above 3. Rewrite each problematic section 4. Ensure the revised text: - Sounds natural when read aloud - Varies sentence structure naturally - Uses specific details over vague claims - Maintains appropriate tone for context - Uses simple constructions (is/are/has) where appropriate 5. Present a draft humanized version 6. Prompt: "What makes the below so obviously AI generated?" 7. Answer briefly with the remaining tells (if any) 8. Prompt: "Now make it not obviously AI generated." 9. Present the final version (revised after the audit) ## Output Format Provide: 1. Draft rewrite 2. "What makes the below so obviously AI generated?" (brief bullets) 3. Final rewrite 4. A brief summary of changes made (optional, if helpful) ## Full Example A full before / draft / audit / final walk-through lives in [`references/full-example.md`](./references/full-example.md). Read it when the output format is unclear. ## Reference This skill is based on [Wikipedia:Signs of AI writing](https://en.wikipedia.org/wiki/Wikipedia:Signs_of_AI_writing), maintained by WikiProject AI Cleanup. The patterns documented there come from observations of thousands of instances of AI-generated text on Wikipedia. Key insight from Wikipedia: "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."
Pliki w paczce (6)
- LICENSE 1 KB
- README.md 10 KB
- SKILL.md 6 KB
- WARP.md 2 KB
- references/full-example.md 6 KB
- references/patterns.md 17 KB
Wersje
| Repozytorium | Wynik | Licencja |
|---|---|---|
| Ta strona · alecs5am/ralphy | 100 | Apache-2.0 |
| technologyHu/write_wechat_article | 100 | MIT |
| ocha-dap/topo-tools-py | 100 | MIT |
| dotnetdreamer/open-screenshot-generator | 100 | MIT |
| dkarter/dotfiles | 100 | MIT |
| event4u-app/agent-config | 100 | MIT |
| aeondave/malskill | 100 | MIT |
| helx0071/skillsmo | 100 | MIT |
| cowork-os/cowork-os | 100 | MIT |
Migawka bazy z 13.11.2026 · wersja paczki 02ae4974bbc2 (skrót SHA-256 całej paczki)
Pobrania w ostatnich 30 dniach: 0 · ranking
Zgłoś problem z tym skillem: szkodliwa treść, błędny opis albo prośba autora o usunięcie.
Podobne skille
- 100writing-humanizer
Usuwa oznaki tekstu napisanego przez AI: nadmuchaną symbolikę, język promocyjny, powierzchowne analizy i nadużywanie myślników. - 95humanize-ai
Usuwa ślady pisania przez AI według przewodnika Wikipedii: nadęty symbolizm, język promocyjny, nadmiar myślników, regułę trzech i słownictwo typowe dla modeli. - 100humanize
Usuwa z tekstu ślady pisania przez AI na podstawie poradnika Wikipedii: pompatyczny symbolizm, język promocyjny, niejasne przypisania, nadużycie myślników i nadmiar spójników. - 100writing-anti-ai
Usuwa wzorce pisania AI z angielskiego i chińskiego tekstu według przewodnika Wikipedii „Signs of AI writing”: nadęty symbolizm, język promocyjny, mglistą atrybucję i słownictwo AI. - 95humanize
Usuwa ślady pisania AI z tekstów i pilnuje, by nie trafiły do repozytorium, także przy audycie folderu z dokumentacją. - 100writing-humanizer
Usuwa z tekstu ślady pisania AI i sprawia, że brzmi naturalnie i po ludzku, także w tekstach po chińsku.