Humanize Mba Text logo

Humanize Mba Text

Community
stephenlzc
humanize-mba-text

去除中文文本中的 AI 生成痕迹,使其更符合中国 MBA 毕业论文的自然写作风格。当用户提到"去 AI 痕迹"、"去除 AI 写作痕迹"、"MBA 论文改写"、"让这段文字更像人写的"、"去除机器感"等关键词时触发。适用于需要降低文本 AI 特征、提升学术写作自然度的场景。

Overview

Publisherstephenlzc
Repositoryhumanize-mba-text-skill
Skill namehumanize-mba-text
Stars
59
Forks
8
Bundled files
48
LicenseMIT
Links
  • Markdown instructions

    A SKILL.md file the model loads on demand, so it only costs tokens when a request actually matches.

  • Works with any LLM

    AI skills are plain Markdown, not provider-specific code, so this works with GPT, Claude, Gemini, Grok, or a local model.

  • 48 bundled files

    Scripts, templates, and references the model can read while it works. Files are read-only and never executed.

  • Open source

    Published by stephenlzc on GitHub. Read the source before you install it.

Installation

Install the Humanize Mba Text AI skill in TypingMind to use it with any LLM, or drop it into another agent that reads SKILL.md.

1

Install in TypingMind

TypingMind installs a skill straight from its GitHub folder — it reads SKILL.md, bundles the resource files, and stores the result locally.

  1. Open the app and go to Plugins → Skills.
  2. Choose "Install from GitHub".
  3. Paste the skill folder URL below and confirm.
    https://github.com/stephenlzc/humanize-mba-text-skill/tree/main
  4. Enable the skill in any chat where you want it available.
Plugins → Skills → Add skill → From GitHub URL, then paste the folder URL and press Continue.
2

Install in another agent

Any agent that reads the Agent Skills format can use this skill — copy the folder into that agent's skills directory.

Claude Code — .claude/skills
git clone --depth 1 https://github.com/stephenlzc/humanize-mba-text-skill.git \
  .claude/skills/humanize-mba-text
Restart Claude Code after copying so it picks up the new skill.

Use it in TypingMind

Enable Humanize Mba Text in any TypingMind chat and the model takes it from there. Its name and description sit in the system prompt, and the moment a request matches, the model loads the full instructions itself — you never invoke it by hand, and it costs no tokens until it is actually used.

The model loads Humanize Mba Text on its own as soon as a request matches it.

Works with any AI model

AI skills are plain Markdown instructions rather than provider-specific code, so Humanize Mba Text is not tied to the model it was written for. Install it once in TypingMind and use it with GPT-5, Claude, Gemini, Grok, DeepSeek, Mistral, Llama, or a local model you run yourself — all on your own API keys.

  • Loaded only when it is needed

    The system prompt carries just the name and description. The instructions are fetched on the first matching request, so an idle skill costs nothing.

  • Switch models mid-chat

    Because the skill is instructions rather than code, changing model does not break it — the next model reads the same SKILL.md.

Skill instructions

This is the SKILL.md content the model loads. Read it before installing — a skill is instructions your model will follow.

去除 AI 写作痕迹 - MBA 论文风格优化

本文档是 manifest:每节只保留导航 / 简介 / 边界,具体内容请读对应 TOML / Markdown 文件。 删繁就简的目的是让 Claude Code 在最小上下文里读到该项目最稳定的入口信息。

任务目标

将带有明显 AI 生成特征的中文文本改写为自然、人类化的 MBA 毕业论文风格文本,符合中国 MBA 毕业论文的学术规范和实践导向。

核心原则

  1. 实践导向:MBA 论文必须来源于企业管理实际,解决具体管理问题,避免纯理论空谈
  2. 小题大做:选题聚焦明确,"小题深做",避免选题过大、过于宽泛
  3. 客观性优先:用事实和数据替代主观评价和空泛论述
  4. 具体化表达:提供具体企业案例、数据支撑、时间节点,避免笼统描述
  5. 数据溯源:所有数据必须注明来源,确保准确性和可信度
  6. 理论支撑:运用 1-2 个相关理论作为分析框架,避免就事论事
  7. 简洁性:删除冗余修饰、重复表述,保留核心信息
  8. 学术规范:符合中国高校 MBA 论文的写作规范和格式要求
  9. 逻辑严密:论证过程清晰,因果关系明确,避免跳跃式推理
  10. 结构合理:正文字数 3 万字以上,章节分配合理,避免节下无目或一节不足一页

识别与改写策略(数据化)

SKILL.md 不再展开具体的识别词表与改写规则——它们以结构化数据的形式存放在本目录下:

类别数据文件内容
内容/语言/结构改写规则references/rules/categories/*.tomlregex 类规则,由 scripts/rule_loader.iter_regex_categories 自动加载
三维协同优化 + 评估标准references/strategy_optimization.toml维度 1/2/3 核心策略、改写流程、量化 / 定性评估表
改写检查清单references/mba_rewrite_checklist.toml4 类检查项 + 硬性量化阈值
章节特定改写规则references/chapter_rewrite_rules.toml5 章节 × 改写点的 before/after 示例
高频 AI 词 / 连接词references/rules/categories/language.tomlai_buzzwords / excessive_connectors 等条目;与 humanizer-academic-zh 模式 10/12/21 交叉对应

使用方式:写正文时按 references/mba_rewrite_checklist.toml 4 大类逐条核对;遇到具体痕迹词按 references/rules/categories/language.toml 词表替换;如果要引用某个章节的特定模式,查 references/chapter_rewrite_rules.toml 的 chapter.section。

改写流程

输入文本
    ↓
[多维度检测] → 生成检测报告(detect_ai_patterns.generate_report)
    ↓
[modify_plan] → 结构化骨架(analyzers.rewrite_planner.build_modify_plan)
    ↓
[反馈生成] → 修改建议(feedback_generator)
    ↓
[自动修复] → 中英文空格等可机械修复项
    ↓
[手动改写] → 按 checklist.toml / chapter_rewrite_rules.toml 逐项处理
    ↓
[质量验证] → 再次跑 detect_patterns,确认 modify_plan 清空

自动化检测与反馈工作流

多方案融合检测

检测层由三类组成,可独立运行也可组合:

  1. 规则匹配检测 (scripts/detect_ai_patterns.py)

    • 加载 references/rules/**/*.toml 的 7 个分类(structure / rhythm_quality / formatting / content / evidence / language / chapter-categories)
    • 走 iter_regex_categories 渐进式加载,可按 group 或 category_id 选载
    • 命中后产出 PatternMatch(pattern_name, pattern_type, line_number, content, suggestion, severity)
  2. 散文统计 + 结构分析 (scripts/analyzers/)

    • 5 个维度:uniform_sentence_length / uniform_paragraph_length / paragraph_edge_template_repeat / paragraph_structure_uniformity / chapter_template_repeat
    • 输出 AnalyzerIssue(analyzer_id, severity, confidence, location, evidence, suggestion) + 统计指标
  3. 语义链分析 (scripts/analyzers/semantic_chain.py)

    • 10 个维度(3 / 4 / 5):三段式 / 作者罗列 / 方法堆叠 / 摘要模板 / 结论回声 / 笼统问题 / 无来源量化 / 宏观叙事 / 证据链 / 问题-对策追踪
    • 跨段 / 跨章节的链状检测,捕捉段内看不到的模式
  4. 融合评分 (scripts/multi_detector.py)

    • FusionEngine.detect 把统计 + 链 + AI 模式三类 detector 合并
    • StatisticalDetector 把散文指标 + 链指标并入 features 与 issues 通道
  5. 按句高风险聚合 (scripts/analyzers/high_risk_annotator.py) 🆕v1.4

    • 把规则匹配 / 散文 / 链三类 issue 按所在句合并成 high_risk_annotations[]
    • 每条标注带 char 偏移、行号、原文、触发规则的 phrase_replacements 与 before_after_example
    • LLM humanizer agent 可逐句消费,无需自己再做一次"哪句有问题"的二次判断

使用方法

bash
# 基础检测(含 modify_plan)
python scripts/detect_ai_patterns.py input.txt --output report.json

# 多维度融合检测
python scripts/multi_detector.py input.txt --output report.md

# 生成修改建议
python scripts/feedback_generator.py detection_result.json --text input.txt --output feedback.md

# 自动修复空格等简单问题
python scripts/feedback_generator.py detection_result.json --text input.txt --apply

检测结果解读

detect_ai_patterns.generate_report 返回 JSON 含:

  • summary.ai_score:0-100,越高 AI 痕迹越重
  • matches:包含规则、散文、链三类问题的统一列表
  • modify_plan ⭐:每条 ModifyEntry 含 location / evidence / rewrite_template / recommended_replacements / target_word_count_range / before_after_example
  • high_risk_annotations 🆕:按句聚合的高风险标注(v1.4),每条 = 一句话 + 该句触发的所有规则 + 短语替换字典 + 真实改写对。LLM humanizer agent 可直接逐句消费
  • chapter_specific_advice:根据检测到的章节类型提供的针对性建议
  • metrics:包含 sentence_cv / paragraph_cv / prose_* / chain_* / evidence_chain_completeness

修改策略见 references/strategy_optimization.toml 的「评估标准」节:高风险 (>70%) 重写、中风险 (40-70%) 针对性优化、低风险 (<40%) 细节润色。

注意事项

  1. 学术诚信:改写时保持学术诚信,不编造数据;所有数据必须有真实来源
  2. 数据核实:对于不确定的引用和数据,建议用户核实原始资料
  3. 理论适配:建议用户根据具体研究问题选择合适的理论框架
  4. 企业授权:如涉及企业内部数据,建议获得企业书面授权或做脱敏处理
  5. 个性化调整:改写后的文本需用户根据具体学校和导师要求调整
  6. 查重控制:改写后建议进行查重,确保复制率 < 15%(详见 references/mba_rewrite_checklist.toml hard_limits.copy_rate)
  7. 功能边界:本 Skill 专注于去除 AI 痕迹和格式规范,不涉及学术内容深度审查
  8. 参考文献:改写不自动生成参考文献,需用户根据实际引用补充完整

补充参考:humanizer-academic-zh(人机分工)

本 Skill 走检测路径——用正则、统计、链分析把 AI 痕迹定位到具体段落。在改写路径上,我们参考同生态的 cangtianhuang/humanizer-academic-zh(其 SYSTEM_PROMPT.md 与 SKILL.md)。两份 Skill 的关系:

维度humanize-mba-text(本 Skill)humanizer-academic-zh
主路径检测 + 量化定位改写 + 润色
规则粒度正则 / CV / 跨段指纹25 类语义模式(每类带"警惕词 + 改写原则")
输出AnalyzerIssue + modify_plan + high_risk_annotations[](按句聚合,含短语替换字典与真实改写对)表格化识别人性化润色版本
强项维度 3-5 的跨段/跨章链检测、量化报告、按句精确定位模式词典丰富、对中文学术语感的精确判断

humanizer-academic-zh 把 AI 模式分成 5 大类共 25 小类。本 Skill 已将其中部分模式并入渐进式加载的 manifest——具体落位:

humanizer-academic-zh 模式本 Skill 的 category_id状态
模式 21 填充短语filler_phrases增强 +3 条正则,覆盖「具有处理……的能力」「不难发现,两人之间存在」「需要指出的是,」
模式 22 过度对冲hedging_overload已有覆盖,本版本不动
模式 23 协作式沟通痕迹chatbot_conversation_residue增强 +2 条正则:「亲爱?的?读者」「如下所示」+「以下是 … 修改/改写/润色 …」
模式 24 知识截止免责knowledge_cutoff_disclaimer已有覆盖,本版本不动
模式 25 谄媚语气sycophantic_praise新增 [[categories]],severity=medium / weight=0.5

所有上述模式的 category_id 都登记在 references/rules/index.toml 的 category_ids 列表里,由 scripts/rule_loader.iter_regex_categories 自动发现。新增模式无须改代码——只要在对应 category 文件里加一行 regex_patterns 项即可。

剩余模式(22 类人文化的句式、连接词密度、否定式排比等)已在原 language.toml 的 19 个 category 里独立实现,不与 humanizer-academic-zh 的分类一一对应;两套体系互补不冲突。

参考文档

自动加载的检测规则

  • references/chinese-paper-humanization-rules.toml — Python rule_loader.load_rules 的默认入口(fallback)
  • references/rules/index.toml — 渐进式加载的 manifest
  • references/rules/categories/{structure, rhythm_quality, formatting, content, evidence, language}.toml — 7 大类 AI 检测规则
  • references/rules/chapter-categories.toml — 章节类型识别
  • references/rules/metrics.toml — 共享度量词表

改写侧策略数据

  • references/strategy_optimization.toml — 三维协同 + 评估标准
  • references/mba_rewrite_checklist.toml — 改写检查清单 + 硬阈值
  • references/chapter_rewrite_rules.toml — 五章节特定改写规则
  • references/strategy_ai_reduction.md — 策略文档(人读版,gitignored 但本地有)
  • references/strategy_plagiarism.md — 策略文档(同上)
  • references/strategy_polishing.md — 策略文档(同上)

配套指南

  • references/ai-writing-patterns.md — AI 写作特征详细列表
  • references/chapter-1-introduction.md 至 references/chapter-5-conclusion.md — 五章节详细写作指南
  • references/format-standards.md — 格式规范

Bundled files

The model reads these on demand while the skill is loaded. They are exposed as readable files and are never executed.

Frequently asked questions

What does the Humanize Mba Text AI skill do?

去除中文文本中的 AI 生成痕迹,使其更符合中国 MBA 毕业论文的自然写作风格。当用户提到"去 AI 痕迹"、"去除 AI 写作痕迹"、"MBA 论文改写"、"让这段文字更像人写的"、"去除机器感"等关键词时触发。适用于需要降低文本 AI 特征、提升学术写作自然度的场景。

Why use Humanize Mba Text on TypingMind?

Because you install it once and use it with any model. Humanize Mba Text is plain Markdown rather than provider-specific code, so the same skill runs on GPT-5, Claude, Gemini, Grok, or a local model — and you can switch model mid-chat without it breaking. TypingMind runs on your own API keys, so you pay providers directly instead of a per-seat subscription, and your skills and chats stay in your own storage.

How do I install Humanize Mba Text in TypingMind?

Open Plugins → Skills → Install from GitHub in TypingMind and paste https://github.com/stephenlzc/humanize-mba-text-skill/tree/main. TypingMind reads its SKILL.md and bundles its files and installs it as a skill you can enable per chat.

Which AI models can use Humanize Mba Text?

Any model you connect in TypingMind. AI skills are plain Markdown instructions rather than provider-specific code, so GPT, Claude, Gemini, Grok, and local models can all load this skill when a request matches it.

How many AI models can I use with Humanize Mba Text?

As many as you like. As long as a model supports skills, you can use Humanize Mba Text with it — GPT, Claude, Gemini, Grok, DeepSeek, Mistral, Llama and more — all on TypingMind with your own API keys.

Is the Humanize Mba Text AI skill free?

Yes. It is published on GitHub by stephenlzc under the MIT license. You only pay your own AI provider for the tokens you use.

What are AI skills?

An AI skill is a reusable instruction bundle that teaches an AI model how to do one specific task. It follows the open Agent Skills format: a SKILL.md file with a name and description, plus any scripts, templates or reference files the model may need. The model reads the instructions only when your request matches the skill, so an installed skill costs nothing until it is used.

How are AI skills different from plugins or MCP servers?

A plugin or MCP server gives a model new tools to call — code that runs somewhere and returns a result. An AI skill gives the model knowledge and process instead: how to approach a task, which steps to follow, what good output looks like. Skills are plain Markdown, so they need no server, no API key and no runtime, and they work with any model.

View all

Set up your own AI workspace now

Get notified about new features and future giveaways by subscribing to our newsletter 👇