antislop by aaaaqwq/agi-super-skills
npx skills add https://github.com/aaaaqwq/agi-super-skills --skill antislop一个全面的 AI 写作模式检测与修复工具。结合了 Wikipedia's Signs of AI Writing 中的模式与高级结构检测,以及一个能实际修复问题的编辑器模式。
星座运势测试:
"这段话是否可能是任何人写给任何人看的?"
如果是,那就是 slop(废话)。就像星座运势——技术上适用于所有人,但与任何人都没有共鸣。
无法通过测试的情况:
能通过测试的情况:
/antislop
[在此处粘贴你的文本]
或者直接要求 Claude 检查文本:
Please run antislop on this: [your text]
广告位招租
在这里展示您的产品或服务
触达数万 AI 开发者,精准高效
这些短语与 AI 关联性极强,仅凭其存在就表明是未经编辑的输出。
| 模式 | 示例 | 修复方法 |
|---|---|---|
| Delve | "Let's delve into..." | 删除或替换为直接陈述 |
| Game-changer | "This game-changing approach..." | 描述实际影响 |
| Revolutionary | "A revolutionary new method..." | 说明它实际做了什么 |
| Unlock potential | "Unlock your potential..." | 完全删除 |
| Leverage (作为动词) | "Leverage these insights..." | 使用 "Use" |
| It's worth noting | "It's worth noting that..." | 直接陈述事实 |
| Moreover/Furthermore | "Moreover, this approach..." | 删除或使用 "Also" |
| Today's digital landscape | "In today's digital landscape..." | 删除 |
| Cutting-edge | "Cutting-edge solutions..." | 删除 |
| Pivotal moment | "Marking a pivotal moment in..." | 说明发生了什么 |
| Tapestry (抽象用法) | "A rich tapestry of influences..." | 删除或具体化 |
| Intricate/intricacies | "The intricacies of..." | 使用 "Details of" 或删除 |
| Showcase (作为动词) | "Showcasing their commitment..." | 使用 "Shows" 或描述发生了什么 |
| Vibrant | "A vibrant community of..." | 删除或使用具体细节 |
| Interplay | "The interplay between X and Y..." | 使用 "How X and Y affect each other" |
| Garner | "Garnering attention from..." | 使用 "Got attention from" 或具体化 |
| Align with | "Aligning with broader trends..." | 说明实际关系 |
研究证据:
过度使用或聚集出现时有问题。
| 模式 | 示例 | 修复方法 |
|---|---|---|
| Here's the thing | 重复使用 | 保留第一个,后续使用变体 |
| At the end of the day | "At the end of the day..." | 删除 |
| The bottom line | "The bottom line is..." | 直接陈述 |
| Let's dive in | "Without further ado, let's dive in" | 删除 |
| Comprehensive and thorough | 成对的形容词 | 选一个 |
| Simple and straightforward | 成对的形容词 | 选一个 |
| In this post, we'll cover | 模板化开头 | 删除 |
| By the end of this article | 承诺式开头 | 删除 |
单独出现没问题,聚集在一起则有问题。
| 模式 | 示例 | 修复方法 |
|---|---|---|
| However/But | 每个段落都这样开头 | 变换过渡词 |
| Firstly/Secondly/Thirdly | 列举要点 | 使用自然流畅的表达 |
| Moving forward | "Moving forward, we'll..." | 删除 |
| Robust/Seamless/Scalable | 企业流行语 | 使用具体术语 |
| Stakeholder | "Key stakeholders..." | 点名或使用 "people" |
---|---|---|---
1 | 重要性夸大 | "marking a pivotal moment in the evolution of..." | "was established in 1989 to collect statistics"
2 | 随意提及知名度 | "cited in NYT, BBC, FT, and The Hindu" | "In a 2024 NYT interview, she argued..."
3 | 肤浅的 -ing 分析 | "symbolizing... reflecting... showcasing..." | 删除或用实际来源扩展
4 | 宣传性语言 | "nestled within the breathtaking region" | "is a town in the Gonder region"
5 | 模糊归因 | "Experts believe it plays a crucial role" | "according to a 2019 survey by..."
6 | 公式化挑战 | "Despite challenges... continues to thrive" | 关于实际挑战的具体事实
7 | 提纲式结论 | "Challenges" 部分以乐观展望结尾 | 删除或替换为实际分析
---|---|---|---
7 | 避免使用系动词 | "serves as... features... boasts..." | "is... has..."
8 | 否定性平行结构 | "It's not just X, it's Y" | 直接陈述观点
9 | 三原则 | "innovation, inspiration, and insights" | 使用自然数量的项目
10 | 同义词循环 | "protagonist... main character... central figure..." | "protagonist" (在意思最清晰时重复)
11 | 虚假范围 | "from the Big Bang to dark matter" | 直接列出主题
12 | 临床式正式用语 | "individuals" / "utilize" / "implement" | "people" / "use" / "do"
---|---|---|---
13 | 过度使用破折号 | "institutions—not the people—yet this continues—" | 使用逗号或句号
14 | 过度使用粗体 | "OKRs , KPIs , BMC " | "OKRs, KPIs, BMC"
15 | 表情符号标题 | "🎯 Goal / 💡 Key Insight / ✅ Action Item" | 删除表情符号
16 | 标题大小写标题 | "Strategic Negotiations And Partnerships" | "Strategic negotiations and partnerships"
17 | 列表成瘾 | 所有内容都变成要点 | 在适当的地方转换为散文
18 | 花引号 | "like this" 而不是 "like this" | 一致使用直引号
19 | 不必要的表格 | 本应是一句话的 3 行表格 | 转换为散文
这些模式绕过了基于短语的检测,但却是主要特征。
三个或更多连续的简短陈述句,以平行结构陈述事实。这是 AI 将要点伪装成散文的方式。
修复前:
The model is impressive. Complex code ships fast. Documentation writes itself. Problems get solved quickly.
修复后:
The model is impressive — complex code ships in a single session, documentation practically writes itself, and problems that would have taken a weekend now take an afternoon.
检测规则: 3 个以上连续的句子,都少于 10 个单词,都是陈述句,遵循平行结构,并且可以变成要点。
每个句子都是 10-15 个单词。简短。有力。令人疲惫。
真正的写作有节奏感——混合使用 5 个单词的句子来制造冲击力,以及 25 个单词的句子来探索含义。
修复前:
This isn't theoretical. It's practical. This isn't a feature. It's a philosophy. It's not about X. It's about Y.
修复后:
Here's how it works in practice: [Just state what it is]
AI 喜欢这种修辞模式。听起来很有力,但浪费了文字来告诉你某物不是什么。
每个章节长度相同。所有段落都是 3-4 句话。AI 没有观点,所以它对所有事物都给予平衡的覆盖。真正的写作反映了优先级。
---|---|---|---
18 | 聊天机器人痕迹 | "I hope this helps! Let me know if..." | 完全删除
19 | 信息不足的免责声明 | "While details are limited in available sources..." | 查找来源或删除
20 | 谄媚的语气 | "Great question! You're absolutely right!" | 直接回应
21 | 奉承式三明治 | "While traditional methods have merit, modern approaches offer..." | 陈述你的实际立场
AI 试图听起来像人,但显得做作:
修复前:
Five services. Five tabs. Five headaches. That got old fast. So I built an MCP server that unifies all of them.
修复后:
I run my newsletter on Kit.com. It's a solid platform, but like most SaaS tools, it means another dashboard, another set of menus to navigate, another context switch.
没有刻意制造的冲击力。没有挖苦。只是描述情况。
内容将作者的成就定位为标题,而不是读者的转变。
修复前:
I shipped 11 MCP servers over the holidays. Here's what I learned.
修复后:
Most developers using Claude Code aren't aware that [observation about the reader's situation]. Here's what's changing...
作者的经验是 证据 ,而不是故事本身。
承诺提供见解但只提供模板结构的标题:
修复方法: 替换为描述性标题,总结实际内容。
---|---|---|---
22 | 填充短语 | "In order to" / "Due to the fact that" | "To" / "Because"
23 | 过度模糊 | "could potentially possibly" | "may"
24 | 通用结论 | "The future looks bright" | 具体计划或事实
| 模式类型 | 分数 |
|---|---|
| 每个第 1 级短语 | +3 |
| 每个第 2 级短语 (重复) | +2 |
| 第 3 级聚集 (章节内 3 个以上) | +2 |
| 未通过星座运势测试 | +5 |
| 断奏式片段泛滥 (每个实例) | +4 |
| 检测到句子均匀性 | +3 |
| 对比句 (每个实例) | +2 |
| 刻意制造个性 | +4 |
| 自我宣传式框架 | +5 |
| 模板标题 (每个实例) | +2 |
分数解读:
此技能是一个 编辑器 ,而不是批评者。检测后:
修复优先级:
要进行审核而不编辑,请明确请求 "audit only"。
## AntiSlop Report
**Horoscope Test:** [PASS/FAIL] - [reason]
**Slop Score:** [X] → [Y] - [Risk Level]
### Fixes Applied
| Location | Before | After |
|----------|--------|-------|
| Line 3 | "Let's delve into the details" | "Here are the details" |
| Line 15 | "Game-changing approach" | "Different approach" |
### Remaining Considerations
- [Any issues requiring human judgment]
### The Core Principle
Your voice is in the specificity, the opinions, the rough edges, and the rhythm. Protect those.
修复前 (AI 风格):
Great question! Here is an essay on this topic. I hope this helps!
AI-assisted coding serves as an enduring testament to the transformative potential of large language models, marking a pivotal moment in the evolution of software development. In today's rapidly evolving technological landscape, these groundbreaking tools—nestled at the intersection of research and practice—are reshaping how engineers ideate, iterate, and deliver.
At its core, the value proposition is clear: streamlining processes, enhancing collaboration, and fostering alignment. It's not just about autocomplete; it's about unlocking creativity at scale.
- 💡 Speed: Code generation is significantly faster.
- 🚀 Quality: Output quality has been enhanced.
- ✅ Adoption: Usage continues to grow.
In conclusion, the future looks bright. Let me know if you'd like me to expand!
修复后 (已修复):
AI coding assistants speed up some tasks. In a 2024 study by Google, developers using Codex completed simple functions 55% faster than a control group, but showed no improvement on debugging or architectural decisions.
The tools are good at boilerplate: config files, test scaffolding, repetitive refactors. They're bad at knowing when they're wrong.
Mira, an engineer at a fintech startup, said she uses Copilot "like autocomplete for boring code" but reviews every line before committing. Jake, a senior dev at a larger company, disabled it after it kept suggesting patterns from a deprecated internal library.
The productivity claims are hard to verify. GitHub says Copilot users "accept 30% of suggestions," but acceptance is not correctness.
随着 AI 模型的发展,模式会过时。在扫描之前,检查 frontmatter 中的 last-refreshed。如果超过 30 天,请先更新。
更新工作流程:
gemini "Fetch these two pages and extract ALL AI writing patterns, phrases, and detection heuristics listed on each. Return as a structured list with pattern name, example, and which page it came from. Pages: https://en.wikipedia.org/wiki/Wikipedia:Signs_of_AI_writing and https://en.wikipedia.org/wiki/Wikipedia:WikiProject_AI_Cleanup" > /tmp/antislop-refresh.txt
# Signs of AI writing - full wikitext
curl -s "https://en.wikipedia.org/w/api.php?action=parse&page=Wikipedia:Signs_of_AI_writing&prop=wikitext&format=json" | python3 -c "
import json, sys
data = json.load(sys.stdin)
print(data['parse']['wikitext']['*'][:30000])
" > /tmp/antislop-signs.txt
# WikiProject AI Cleanup
curl -s "https://en.wikipedia.org/w/api.php?action=parse&page=Wikipedia:WikiProject_AI_Cleanup&prop=wikitext&format=json" | python3 -c "
import json, sys
data = json.load(sys.stdin)
print(data['parse']['wikitext']['*'][:30000])
" > /tmp/antislop-cleanup.txt
last-refreshed 日期不要添加重复项。 许多 Wikipedia 模式在此处已以不同名称涵盖。只添加代表真正新检测信号的模式。
AI slop 不是关于单个词汇——而是关于模式。
一个 "moreover" 不会让内容变成 AI 生成的。但是 "moreover" + "it's worth noting" + "delve into" + 均匀的句子 + 表情符号标题 = 明显的 slop。
目标是写出听起来像一个具有特定观点的特定人类,而不是一个非常有礼貌、不想冒犯任何人的委员会。
每周安装数
1
代码仓库
GitHub 星标数
11
首次出现
1 天前
安全审计
安装于
zencoder1
amp1
cline1
openclaw1
opencode1
cursor1
A comprehensive AI writing pattern detector and fixer. Combines patterns from Wikipedia's Signs of AI Writing with advanced structural detection and an editor mode that actually fixes problems.
The Horoscope Test:
"Could anyone have written this, for anyone?"
If yes, it's slop. Like a horoscope — technically applicable to everyone, resonant with no one.
What fails:
What passes:
/antislop
[paste your text here]
Or ask Claude to check text directly:
Please run antislop on this: [your text]
These phrases are so strongly associated with AI that their presence alone suggests unedited output.
| Pattern | Example | Fix |
|---|---|---|
| Delve | "Let's delve into..." | Remove or replace with direct statement |
| Game-changer | "This game-changing approach..." | Describe the actual impact |
| Revolutionary | "A revolutionary new method..." | State what it actually does |
| Unlock potential | "Unlock your potential..." | Remove entirely |
| Leverage (as verb) | "Leverage these insights..." | "Use" |
| It's worth noting | "It's worth noting that..." | Just state the thing |
| Moreover/Furthermore | "Moreover, this approach..." | Remove or use "Also" |
| Today's digital landscape | "In today's digital landscape..." | Remove |
| Cutting-edge | "Cutting-edge solutions..." |
Research evidence:
Problematic when overused or clustered.
| Pattern | Example | Fix |
|---|---|---|
| Here's the thing | Used repeatedly | Keep first, vary subsequent |
| At the end of the day | "At the end of the day..." | Remove |
| The bottom line | "The bottom line is..." | Just state it |
| Let's dive in | "Without further ado, let's dive in" | Remove |
| Comprehensive and thorough | Paired adjectives | Pick one |
| Simple and straightforward | Paired adjectives | Pick one |
| In this post, we'll cover | Template opening | Remove |
| By the end of this article | Promise opener | Remove |
Fine individually, problematic together.
| Pattern | Example | Fix |
|---|---|---|
| However/But | Every paragraph starts this way | Vary transitions |
| Firstly/Secondly/Thirdly | Enumerated points | Use natural flow |
| Moving forward | "Moving forward, we'll..." | Remove |
| Robust/Seamless/Scalable | Corporate buzzwords | Use specific terms |
| Stakeholder | "Key stakeholders..." | Name them or say "people" |
---|---|---|---
1 | Significance inflation | "marking a pivotal moment in the evolution of..." | "was established in 1989 to collect statistics"
2 | Notability name-dropping | "cited in NYT, BBC, FT, and The Hindu" | "In a 2024 NYT interview, she argued..."
3 | Superficial -ing analyses | "symbolizing... reflecting... showcasing..." | Remove or expand with actual sources
4 | Promotional language | "nestled within the breathtaking region" | "is a town in the Gonder region"
5 | Vague attributions | "Experts believe it plays a crucial role" | "according to a 2019 survey by..."
6 | Formulaic challenges | "Despite challenges... continues to thrive" | Specific facts about actual challenges
7 | Outline-like conclusions | "Challenges" section ending with optimistic outlook | Remove or replace with actual analysis
---|---|---|---
7 | Copula avoidance | "serves as... features... boasts..." | "is... has..."
8 | Negative parallelisms | "It's not just X, it's Y" | State the point directly
9 | Rule of three | "innovation, inspiration, and insights" | Use natural number of items
10 | Synonym cycling | "protagonist... main character... central figure..." | "protagonist" (repeat when clearest)
11 | False ranges | "from the Big Bang to dark matter" | List topics directly
12 | Clinical formality | "individuals" / "utilize" / "implement" | "people" / "use" / "do"
---|---|---|---
13 | Em dash overuse | "institutions—not the people—yet this continues—" | Use commas or periods
14 | Boldface overuse | "OKRs , KPIs , BMC " | "OKRs, KPIs, BMC"
15 | Emoji headers | "🎯 Goal / 💡 Key Insight / ✅ Action Item" | Remove emojis
16 | Title Case Headings | "Strategic Negotiations And Partnerships" | "Strategic negotiations and partnerships"
17 | List addiction | Everything becomes bullets | Convert to prose where appropriate
18 | Curly quotes | "like this" instead of "like this" | Use straight quotes consistently
19 | Unnecessary tables | 3-row table that should be a sentence | Convert to prose
These bypass phrase-based detection but are major tells.
Three or more consecutive short declarative sentences stating facts in parallel structure. AI's version of bullets pretending to be prose.
Before:
The model is impressive. Complex code ships fast. Documentation writes itself. Problems get solved quickly.
After:
The model is impressive — complex code ships in a single session, documentation practically writes itself, and problems that would have taken a weekend now take an afternoon.
Detection rule: 3+ consecutive sentences that are all under 10 words, all declarative, following parallel structure, and could be bullet points.
Every sentence 10-15 words. Short. Punchy. Exhausting.
Real writing has rhythm — mix 5-word sentences for impact with 25-word sentences that explore implications.
Before:
This isn't theoretical. It's practical. This isn't a feature. It's a philosophy. It's not about X. It's about Y.
After:
Here's how it works in practice: [Just state what it is]
AI loves this rhetorical pattern. It sounds punchy but wastes words telling you what something isn't.
Every section same length. All paragraphs 3-4 sentences. AI doesn't have opinions, so it gives balanced coverage to everything. Real writing reflects priorities.
---|---|---|---
18 | Chatbot artifacts | "I hope this helps! Let me know if..." | Remove entirely
19 | Cutoff disclaimers | "While details are limited in available sources..." | Find sources or remove
20 | Sycophantic tone | "Great question! You're absolutely right!" | Respond directly
21 | Flattery sandwiches | "While traditional methods have merit, modern approaches offer..." | State your actual position
AI trying to sound human but coming across as performative:
Before:
Five services. Five tabs. Five headaches. That got old fast. So I built an MCP server that unifies all of them.
After:
I run my newsletter on Kit.com. It's a solid platform, but like most SaaS tools, it means another dashboard, another set of menus to navigate, another context switch.
No manufactured punch. No snark. Just describes the situation.
Content positioning author's accomplishments as the headline instead of reader's transformation.
Before:
I shipped 11 MCP servers over the holidays. Here's what I learned.
After:
Most developers using Claude Code aren't aware that [observation about the reader's situation]. Here's what's changing...
The author's experience is evidence , not the story.
Headers that promise insight but deliver template structure:
Fix: Replace with descriptive headers that summarize the actual content.
---|---|---|---
22 | Filler phrases | "In order to" / "Due to the fact that" | "To" / "Because"
23 | Excessive hedging | "could potentially possibly" | "may"
24 | Generic conclusions | "The future looks bright" | Specific plans or facts
| Pattern Type | Points |
|---|---|
| Each Tier 1 phrase | +3 |
| Each Tier 2 phrase (repeated) | +2 |
| Tier 3 cluster (3+ in section) | +2 |
| Failed horoscope test | +5 |
| Staccato fragment spam (per instance) | +4 |
| Sentence uniformity detected | +3 |
| Comparator sentences (per instance) | +2 |
| Manufactured personality | +4 |
| Self-promotional framing | +5 |
| Template headers (per instance) | +2 |
Score interpretation:
This skill is an editor , not a critic. After detection:
Fix priority:
To audit without editing, explicitly request "audit only."
## AntiSlop Report
**Horoscope Test:** [PASS/FAIL] - [reason]
**Slop Score:** [X] → [Y] - [Risk Level]
### Fixes Applied
| Location | Before | After |
|----------|--------|-------|
| Line 3 | "Let's delve into the details" | "Here are the details" |
| Line 15 | "Game-changing approach" | "Different approach" |
### Remaining Considerations
- [Any issues requiring human judgment]
### The Core Principle
Your voice is in the specificity, the opinions, the rough edges, and the rhythm. Protect those.
Before (AI-sounding):
Great question! Here is an essay on this topic. I hope this helps!
AI-assisted coding serves as an enduring testament to the transformative potential of large language models, marking a pivotal moment in the evolution of software development. In today's rapidly evolving technological landscape, these groundbreaking tools—nestled at the intersection of research and practice—are reshaping how engineers ideate, iterate, and deliver.
At its core, the value proposition is clear: streamlining processes, enhancing collaboration, and fostering alignment. It's not just about autocomplete; it's about unlocking creativity at scale.
- 💡 Speed: Code generation is significantly faster.
- 🚀 Quality: Output quality has been enhanced.
- ✅ Adoption: Usage continues to grow.
In conclusion, the future looks bright. Let me know if you'd like me to expand!
After (Fixed):
AI coding assistants speed up some tasks. In a 2024 study by Google, developers using Codex completed simple functions 55% faster than a control group, but showed no improvement on debugging or architectural decisions.
The tools are good at boilerplate: config files, test scaffolding, repetitive refactors. They're bad at knowing when they're wrong.
Mira, an engineer at a fintech startup, said she uses Copilot "like autocomplete for boring code" but reviews every line before committing. Jake, a senior dev at a larger company, disabled it after it kept suggesting patterns from a deprecated internal library.
The productivity claims are hard to verify. GitHub says Copilot users "accept 30% of suggestions," but acceptance is not correctness.
Patterns go stale as AI models evolve. Before scanning, check last-refreshed in frontmatter. If >30 days old, refresh first.
Refresh workflow:
gemini "Fetch these two pages and extract ALL AI writing patterns, phrases, and detection heuristics listed on each. Return as a structured list with pattern name, example, and which page it came from. Pages: https://en.wikipedia.org/wiki/Wikipedia:Signs_of_AI_writing and https://en.wikipedia.org/wiki/Wikipedia:WikiProject_AI_Cleanup" > /tmp/antislop-refresh.txt
# Signs of AI writing - full wikitext
curl -s "https://en.wikipedia.org/w/api.php?action=parse&page=Wikipedia:Signs_of_AI_writing&prop=wikitext&format=json" | python3 -c "
import json, sys
data = json.load(sys.stdin)
print(data['parse']['wikitext']['*'][:30000])
" > /tmp/antislop-signs.txt
# WikiProject AI Cleanup
curl -s "https://en.wikipedia.org/w/api.php?action=parse&page=Wikipedia:WikiProject_AI_Cleanup&prop=wikitext&format=json" | python3 -c "
import json, sys
data = json.load(sys.stdin)
print(data['parse']['wikitext']['*'][:30000])
" > /tmp/antislop-cleanup.txt
3. Read the output and diff against patterns already in this skill
4. For genuinely new patterns not already covered:
* Classify into Tier 1/2/3 based on how strongly they signal AI
* Add to the appropriate table with example and fix
* Update the pattern count in the overview
5. Update last-refreshed date in frontmatter
6. Report what was added (if anything)
Don't add duplicates. Many Wikipedia patterns are already covered here under different names. Only add patterns that represent genuinely new detection signals.
AI slop isn't about individual words — it's about patterns.
One "moreover" doesn't make content AI-generated. But "moreover" + "it's worth noting" + "delve into" + uniform sentences + emoji headers = obvious slop.
The goal is writing that sounds like a specific human with specific opinions, not a very polite committee trying not to offend anyone.
Weekly Installs
1
Repository
GitHub Stars
11
First Seen
1 day ago
Security Audits
Gen Agent Trust HubFailSocketFailSnykWarn
Installed on
zencoder1
amp1
cline1
openclaw1
opencode1
cursor1
内容创作指南:跨渠道营销内容模板与SEO优化技巧
1,300 周安装
| Remove |
| Pivotal moment | "Marking a pivotal moment in..." | State what happened |
| Tapestry (abstract) | "A rich tapestry of influences..." | Remove or be specific |
| Intricate/intricacies | "The intricacies of..." | "Details of" or remove |
| Showcase (as verb) | "Showcasing their commitment..." | "Shows" or describe what happened |
| Vibrant | "A vibrant community of..." | Remove or use specific detail |
| Interplay | "The interplay between X and Y..." | "How X and Y affect each other" |
| Garner | "Garnering attention from..." | "Got attention from" or be specific |
| Align with | "Aligning with broader trends..." | State the actual relationship |