Loading skill
Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Remove AI-like features from text to make it more natural and human-like (去除AI生成文本的AI特征，使文本更自然、更接近人类写作)
.claude/skills/deai-skill/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-11 | ✗→✓ | ▲ Improved | — | — |
| case-09 | ✗→✓ | ▲ Improved | — | — |
| case-07 | ✗→✓ | ▲ Improved | — | — |
| case-18 | ✗→✓ | ▲ Improved | — | — |
| case-04 | ✗→✓ | ▲ Improved | — | — |
【中文】 本技能是一个专业的AI文本去AI味处理工具，用于去除AI生成文本中的典型AI特征，使文本看起来更加自然、接近人类写作风格。该技能支持中文和英文两种语言，提供多种处理风格，适用于各种文本处理场景。
核心功能包括：
【English】 This skill is a professional AI text humanization tool designed to remove typical AI features from AI-generated text, making it appear more natural and closer to human writing style. The skill supports both Chinese and English languages, provides multiple processing styles, and is suitable for various text processing scenarios.
Core features include:
【中文】
pythonfrom deai_skill import DeAIProcessor # 创建处理器实例 processor = DeAIProcessor() # 处理中文文本 chinese_text = "综上所述，我们可以看出人工智能技术正在改变我们的生活方式。" humanized_text = processor.process(chinese_text, language='zh') print(humanized_text) # 处理英文文本 english_text = "In conclusion, it is important to note that artificial intelligence technology is transforming our way of life." humanized_text = processor.process(english_text, language='en') print(humanized_text)
bash# 处理单个文件 python deai_skill.py --input input.txt --output output.txt --language zh # 批量处理目录 python deai_skill.py --input_dir ./texts --output_dir ./humanized --language en # 指定风格 python deai_skill.py --input input.txt --output output.txt --language zh --style casual
【English】
pythonfrom deai_skill import DeAIProcessor # Create processor instance processor = DeAIProcessor() # Process Chinese text chinese_text = "综上所述，我们可以看出人工智能技术正在改变我们的生活方式。" humanized_text = processor.process(chinese_text, language='zh') print(humanized_text) # Process English text english_text = "In conclusion, it is important to note that artificial intelligence technology is transforming our way of life." humanized_text = processor.process(english_text, language='en') print(humanized_text)
bash# Process single file python deai_skill.py --input input.txt --output output.txt --language zh # Batch process directory python deai_skill.py --input_dir ./texts --output_dir ./humanized --language en # Specify style python deai_skill.py --input input.txt --output output.txt --language zh --style casual
【中文】
本技能提供三种处理风格，可根据不同场景选择：
技能使用 rules.json 配置文件来定义转换规则和特征词列表。主要配置项包括：
chinese_ai_features: 中文AI特征词列表transition_words: 中文过渡词（如"综上所述"、"值得注意的是"）formal_words: 中文正式词汇（如"具有"、"体现出"）idioms: 中文成语（如"举足轻重"、"不可或缺"）english_ai_features: 英文AI特征词列表transition_words: 英文过渡词（如"furthermore"、"moreover"）formal_words: 英文正式词汇（如"possess"、"exhibit"）idioms: 英文常用短语（如"play a crucial role"、"of great importance"）conversion_rules: 文本转换规则transition_replacements: 过渡词替换规则formal_word_replacements: 正式词汇替换规则sentence_patterns: 句式模式转换规则colloquial_expressions: 口语化表达endings: 句尾语气词fillers: 填充词uncertainties: 不确定性表达【English】
This skill provides three processing styles that can be selected according to different scenarios:
The skill uses the rules.json configuration file to define transformation rules and feature word lists. Main configuration items include:
chinese_ai_features: List of Chinese AI feature wordstransition_words: Chinese transition words (e.g., "综上所述", "值得注意的是")formal_words: Chinese formal vocabulary (e.g., "具有", "体现出")idioms: Chinese idioms (e.g., "举足轻重", "不可或缺")english_ai_features: List of English AI feature wordstransition_words: English transition words (e.g., "furthermore", "moreover")formal_words: English formal vocabulary (e.g., "possess", "exhibit")idioms: English common phrases (e.g., "play a crucial role", "of great importance")conversion_rules: Text transformation rulestransition_replacements: Transition word replacement rulesformal_word_replacements: Formal word replacement rulessentence_patterns: Sentence pattern transformation rulescolloquial_expressions: Colloquial expressionsendings: Sentence ending particlesfillers: Filler wordsuncertainties: Uncertainty expressions【中文】
输入文本：
综上所述，通过深入分析可以看出，人工智能技术在教育领域的应用具有巨大的潜力和广阔的前景。我们应该认识到，这一技术能够为学生提供个性化学习体验，同时也需要关注其可能带来的挑战。处理结果（自然风格）：
总的来说，从分析中能看出来，AI技术在教育这块确实挺有潜力，前景也不错。它能给学生提供个性化的学习体验啊，不过咱们也得注意可能会遇到的一些问题。AI特征分析：
输入文本：
In conclusion, after thorough analysis, it becomes evident that artificial intelligence technology possesses immense potential and promising prospects in the field of education. We should recognize that this technology can provide personalized learning experiences for students, while also paying attention to potential challenges that may arise.处理结果（自然风格）：
Overall, looking at the analysis, AI technology clearly has huge potential in education. It can give students personalized learning experiences, but we also need to be aware of the challenges that might come up.AI特征分析：
原文：
这一问题具有多个方面的考虑因素，需要我们进行深入的分析和研究。值得注意的是，这一技术方案展现出明显的优势。自然风格：
这一问题具有多个方面的考虑因素，需要我们进行深入的分析和研究。值得注意的是，这一技术方案展现出明显的优势呀。口语风格：
这一问题具有多个方面的考虑因素，需要我们进行深入的分析和研究。有一点要提醒，这一技术方案展现出明显的优势呢。正式风格：
这一问题具有多个方面的考虑因素，需要我们进行深入的分析和研究。得注意一下，这一技术方案展现出明显的优势。【English】
Input Text:
综上所述，通过深入分析可以看出，人工智能技术在教育领域的应用具有巨大的潜力和广阔的前景。我们应该认识到，这一技术能够为学生提供个性化学习体验，同时也需要关注其可能带来的挑战。Processing Result (Natural Style):
总的来说，从分析中能看出来，AI技术在教育这块确实挺有潜力，前景也不错。它能给学生提供个性化的学习体验啊，不过咱们也得注意可能会遇到的一些问题。AI Feature Analysis:
Input Text:
In conclusion, after thorough analysis, it becomes evident that artificial intelligence technology possesses immense potential and promising prospects in the field of education. We should recognize that this technology can provide personalized learning experiences for students, while also paying attention to potential challenges that may arise.Processing Result (Natural Style):
Overall, looking at the analysis, AI technology clearly has huge potential in education. It can give students personalized learning experiences, but we also need to be aware of the challenges that might come up.AI Feature Analysis:
Original Text:
这一问题具有多个方面的考虑因素，需要我们进行深入的分析和研究。值得注意的是，这一技术方案展现出明显的优势。Natural Style:
这一问题具有多个方面的考虑因素，需要我们进行深入的分析和研究。值得注意的是，这一技术方案展现出明显的优势呀。Casual Style:
这一问题具有多个方面的考虑因素，需要我们进行深入的分析和研究。有一点要提醒，这一技术方案展现出明显的优势呢。Formal Style:
这一问题具有多个方面的考虑因素，需要我们进行深入的分析和研究。得注意一下，这一技术方案展现出明显的优势。【中文】
formal风格，保留更多专业术语和正式表达casual风格，增加口语化程度natural风格，平衡专业性和自然度Q: 处理后的文本为什么还有一些AI特征？ A: 本技能采用概率性的规则应用，为了保持文本的多样性，并非所有特征都会被替换。如果需要更强的处理效果，可以尝试：
Q: 能否处理其他语言的文本？ A: 当前版本仅支持中文和英文。其他语言的支持正在开发中。
Q: 处理速度如何？ A: 处理速度约为每秒1000字符。对于批量处理，建议使用批量处理接口以获得更好的性能。
Q: 如何自定义规则？ A: 编辑rules.json文件，可以添加、修改或删除各种规则。详细说明请参考配置文件注释。
【English】
formal style to preserve more technical terms and formal expressionscasual style to increase colloquialismnatural style to balance professionalism and naturalnessQ: Why does the processed text still have some AI features? A: This skill uses probabilistic rule application, and not all features will be replaced to maintain text diversity. If stronger processing results are needed, you can try:
Q: Can it process texts in other languages? A: The current version only supports Chinese and English. Support for other languages is under development.
Q: What is the processing speed? A: The processing speed is approximately 1000 characters per second. For batch processing, it is recommended to use the batch processing interface for better performance.
Q: How can I customize rules? A: Edit the rules.json file, where you can add, modify, or delete various rules. For detailed instructions, please refer to the configuration file comments.
【中文】
【English】
【中文】 本项目采用 MIT 许可证。详见 LICENSE 文件。
【English】 This project is licensed under the MIT License. See the LICENSE file for details.
【中文】
【English】
【中文】 感谢所有为这个项目做出贡献的开发者和用户。特别感谢OpenClaw团队提供的支持。
【English】 Thanks to all developers and users who have contributed to this project. Special thanks to the OpenClaw team for their support.
| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-11 | fail→pass | — | — | — | — | — | — | — | — | — | — | — | — |
case-06 | pass→pass | — | — | — | — | — | — | — | — | — | — | — | — |
case-09 | fail→pass | — | — | — | — | — | — | — | — | — | — | — | — |
case-07 | fail→pass | — | — | — | — | — | — | — | — | — | — | — | — |
case-18 | fail→pass | — | — | — | — | — | — | — | — | — | — | — | — |
case-04 | fail→pass | — | — | — | — | — | — | — | — | — | — | — | — |
case-21 | fail→fail | — | — | — | — | — | — | — | — | — | — | — | — |
case-03 | fail→pass | — | — | — | — | — | — | — | — | — | — | — | — |
case-08 | fail→pass | — | — | — | — | — | — | — | — | — | — | — | — |
case-10 | fail→pass | — | — | — | — | — | — | — | — | — | — | — | — |
case-02 | fail→fail | — | — | — | — | — | — | — | — | — | — | — | — |
case-16 | pass→pass | — | — | — | — | — | — | — | — | — | — | — | — |
case-12 | fail→pass | — | — | — | — | — | — | — | — | — | — | — | — |
case-13 | pass→pass | — | — | — | — | — | — | — | — | — | — | — | — |
case-22 | fail→fail | — | — | — | — | — | — | — | — | — | — | — | — |
case-14 | fail→pass | — | — | — | — | — | — | — | — | — | — | — | — |
case-20 | fail→fail | — | — | — | — | — | — | — | — | — | — | — | — |
case-15 | fail→fail | — | — | — | — | — | — | — | — | — | — | — | — |
case-19 | fail→pass | — | — | — | — | — | — | — | — | — | — | — | — |
case-05 | fail→pass | — | — | — | — | — | — | — | — | — | — | — | — |
case-01 | fail→pass | — | — | — | — | — | — | — | — | — | — | — | — |
case-17 | fail→fail | — | — | — | — | — | — | — | — | — | — | — | — |
case-23 | fail→fail | — | — | — | — | — | — | — | — | — | — | — | — |
DecimalAI ran this skill against gemini-3.6-flash twice over the same eval suite — once with the skill loaded and once without — and compared the two runs case by case. 23 cases were attempted. The headline lift of +57 percentage points is the difference between those two pass rates over the 23 comparable cases.
The per-case answers from this run were removed by the retention sweep, so the case table below shows the verdicts without the text either arm produced. The counts above were recorded at the time and are unaffected. Answers are now kept for 180 days.
Other measured skills in the registry, with their headline benchmark lift.