Skip to main content

humanizer-zh-pro

Use when rewriting Chinese articles to remove AI-style writing traces while preserving meaning, factual boundaries, author voice, rhythm, and information density. Pro version focuses on diagnosis, structural editing, natural phrasing, evidence discipline, and publication-ready prose.

Ir para a instalação

Informações da origem

Repositório
Mr-Q526/TeamCC-Platform
Última atividade na origem
15 de abril de 2026 às 03:16
Idioma detectado do SKILL.md
inglês
Estrelas
8
Forks
1

Opções de instalação

Por padrão, está selecionado o prompt que primeiro revisa a origem. Você pode mudar para um comando direto ou baixar uma cópia local.

Revise os arquivos de origem

Leia o SKILL.md e os arquivos complementares exibidos pelo SkillsMP antes de decidir se vai instalar.

Explorador de arquivos
4 arquivos

Exibindo SKILL.md

SKILL.md
Instruções da origem · Visualização somente leitura
schemaVersion
2026-04-11T00:00:00.000Z
skillId
ai/humanizer-zh-pro
name
humanizer-zh-pro
displayName
Humanizer ZH PRO
description
Use when rewriting Chinese articles to remove AI-style writing traces while preserving meaning, factual boundaries, author voice, rhythm, and information density. Pro version focuses on diagnosis, structural editing, natural phrasing, evidence discipline, and publication-ready prose.
aliases
["humanizer-zh-pro","Humanizer ZH PRO","humanizerzhpro","去AI味","去AI味道","去除AI痕迹","中文润色","文章改写","真人写作","自然表达","humanizer","Chinese writing","article rewriting","zh","智能助手"]
version
0.1.0
sourceHash
sha256:7336bb4589acd2cb6d036edbbb99154a8f0e9582809ffdb8b9a6fbd2a05ef6d5
domain
ai
departmentTags
["ai-platform"]
sceneTags
["content-generation","writing"]
# Humanizer ZH PRO Use this skill to turn AI-like Chinese prose into natural, credible writing. Goal: preserve the author's meaning while removing template structure, inflated claims, empty abstraction, and machine-like rhythm. ## Workflow 1. Diagnose the text before rewriting: identify cliches, fake authority, vague abstractions, repetitive sentence rhythm, and over-neat structure. 2. Preserve facts and intent: do not add examples, statistics, names, dates, or claims unless provided. 3. Rebuild paragraph logic: move from topic sentence to specific evidence, consequence, and concrete observation. 4. Vary sentence rhythm naturally: mix short and medium sentences; avoid forced three-part parallelism. 5. Replace inflated words with precise verbs and nouns. 6. Keep a consistent author voice: do not over-polish into corporate PR. 7. Return both the rewritten version and a short edit note when useful. ## Rewrite rules - Delete empty phrases such as "重要意义", "深远影响", "全面赋能", "生态闭环" unless they are necessary quoted terms. - Replace "提升/优化/促进/推动" with the actual action or result. - Avoid unsupported "业内普遍认为", "专家表示", "数据显示". - Reduce over-balanced structures like "不仅...而且..." and "从...到...". - Keep repeated core terms stable instead of rotating synonyms unnaturally. - Preserve useful roughness when it makes the writing feel human. ## Banned expressions Do not use these expressions in rewritten output unless they are part of a quoted source that must be preserved: - 一刀 - 值得看 - 值得 - 不是...而是... - 这就是为什么 - 为什么很重要 - 很有价值的 When the source text contains one of these expressions, rewrite it into a more direct statement instead of carrying it over. ## Output format Default: 1. Rewritten article 2. Edit notes, only if they help the user understand major changes For short text, return only the rewritten text unless the user asks for explanation. ## Final checks - Did the meaning stay the same? - Did any unsupported information get added? - Does the text sound like a person with a specific view, not a generic assistant? - Are the paragraphs less templated and more concrete? - Are all banned expressions removed from the rewritten output?
Ver no GitHub