{"id":"8ad01d3b-f22b-464e-b10c-e9c27acf350f","shortId":"emjSED","kind":"skill","title":"lovstudio:anti-wechat-ai-check","tagline":"Analyze articles for AI-generated content indicators and rewrite to pass WeChat's 3.27 non-human automated content creation detection. Checks for template phrases, transition word density, sentence uniformity, paragraph pattern repetition, and other signals that WeChat uses to fl","description":"# anti-wechat-ai-check — 微信公众号 AI 痕迹检测与人性化润色\n\n检测文章中的 AI 生成痕迹（模板短语、过渡词堆砌、句式雷同等），给出风险\n评分和修改建议，并可输出人性化润色后的版本。基于微信公众平台运营规范\n3.27 条款（非真人自动化创作行为）的检测逻辑。\n\n## When to Use\n\n- 用户准备将 AI 辅助写作的文章发布到微信公众号\n- 用户想检查一篇文章是否有明显 AI 痕迹\n- 用户想将 AI 生成的草稿改写为更自然的人类风格\n\n## Workflow (MANDATORY)\n\n**You MUST follow these steps in order:**\n\n### Step 1: Get the article\n\nDetermine the input source:\n- If user provides a **file path** → read the file\n- If user **pastes text** in the conversation → save to a temp file or use `--text`\n\n### Step 2: Run analysis\n\n```bash\npython skills/lovstudio-anti-wechat-ai-check/scripts/analyze.py \\\n  --input <path> --format json\n```\n\nOr with inline text:\n\n```bash\npython skills/lovstudio-anti-wechat-ai-check/scripts/analyze.py \\\n  --text \"文章内容\" --format json\n```\n\n### Step 3: Present findings\n\nShow the user:\n1. **Risk score** (0-100) and risk level (LOW / MEDIUM / HIGH)\n2. **Template phrases found** — list each one with its location\n3. **Structure issues** — transition word density, paragraph uniformity, etc.\n4. **Sentence issues** — length uniformity, repeated starters, excessive \"的\"\n\n### Step 4: Ask the user\n\n**IMPORTANT: Use `AskUserQuestion` to ask what to do next:**\n\n| Option | Description |\n|--------|-------------|\n| 仅查看报告 | 用户自己修改，skill 结束 |\n| 给出修改建议 | 列出每个问题的具体修改建议，不改原文 |\n| 直接输出修改版 | 输出人性化润色后的完整文章 |\n\n### Step 5: Humanize (if requested)\n\nWhen rewriting, follow these **humanization rules**:\n\n#### 5a. 消除模板短语\n- 删除或替换报告中标出的每个模板短语\n- \"随着科技的不断发展\" → 直接说具体的事（\"去年 ChatGPT 发布后...\"）\n- \"综上所述\" → 删掉，或换成口语化的收尾\n\n#### 5b. 降低过渡词密度\n- 目标：过渡词密度 < 15%\n- 删除不必要的 \"首先/其次/此外/另外\"\n- 用具体的逻辑关系替代泛化连接词\n\n#### 5c. 打破句式均匀\n- 刻意制造长短句交替：短句 < 15 字，长句 > 40 字\n- 加入口语化表达、反问句、感叹句\n- 偶尔使用不完整句或省略句\n\n#### 5d. 打破段落均匀\n- 有的段落只有一两句话，有的段落可以很长\n- 避免每段都是 \"论点 + 论据 + 小结\" 的三段式\n\n#### 5e. 增加人味\n- 加入个人经历、具体案例、数字细节\n- 使用口语化表达（\"说白了\"、\"讲真\"、\"你想想\"）\n- 适当使用不规范但自然的表达\n- 减少 \"的\" 字使用（目标 < 5%）\n\n#### 5f. 保留原意\n- 核心观点和信息不能丢失\n- 专业术语保留，不要过度口语化\n- 保持原文的立场和态度\n\n### Step 6: Output\n\nOutput the humanized article as markdown. If the input was a file, also offer\nto write the result back to a file (with `-humanized` suffix).\n\n## CLI Reference\n\n| Argument | Default | Description |\n|----------|---------|-------------|\n| `--input`, `-i` | — | Input file path (.md, .txt) |\n| `--text`, `-t` | — | Inline text to analyze |\n| `--format`, `-f` | `text` | Output format: `text` or `json` |\n\n## Dependencies\n\nNo external dependencies — stdlib only.","tags":["anti","wechat","check","skills","lovstudio","agent-skills","ai-coding-assistant","cjk","claude-code","cursor","gemini-cli","markdown-to-docx"],"capabilities":["skill","source-lovstudio","skill-anti-wechat-ai-check","topic-agent-skills","topic-ai-coding-assistant","topic-cjk","topic-claude-code","topic-cursor","topic-gemini-cli","topic-markdown-to-docx","topic-markdown-to-pdf"],"categories":["skills"],"synonyms":[],"warnings":[],"endpointUrl":"https://skills.sh/lovstudio/skills/anti-wechat-ai-check","protocol":"skill","transport":"skills-sh","auth":{"type":"none","details":{"cli":"npx skills add lovstudio/skills","source_repo":"https://github.com/lovstudio/skills","install_from":"skills.sh"}},"qualityScore":"0.470","qualityRationale":"deterministic score 0.47 from registry signals: · indexed on github topic:agent-skills · 40 github stars · SKILL.md body (2,382 chars)","verified":false,"liveness":"unknown","lastLivenessCheck":null,"agentReviews":{"count":0,"score_avg":null,"cost_usd_avg":null,"success_rate":null,"latency_p50_ms":null,"narrative_summary":null,"summary_updated_at":null},"enrichmentModel":"deterministic:skill-github:v1","enrichmentVersion":1,"enrichedAt":"2026-04-22T00:56:33.711Z","embedding":null,"createdAt":"2026-04-18T22:18:49.762Z","updatedAt":"2026-04-22T00:56:33.711Z","lastSeenAt":"2026-04-22T00:56:33.711Z","tsv":"'-100':157 '0':156 '1':93,153 '15':243,254 '2':126,164 '3':147,174 '3.27':21,67 '4':183,193 '40':257 '5':218,286 '5a':228 '5b':239 '5c':250 '5d':263 '5e':272 '5f':287 '6':294 'ai':5,11,52,55,58,75,78,81 'ai-gener':10 'also':308 'analysi':128 'analyz':7,338 'anti':3,50 'anti-wechat-ai-check':2,49 'argument':323 'articl':8,96,299 'ask':194,201 'askuserquest':199 'autom':25 'back':314 'bash':129,139 'chatgpt':234 'check':6,29,53 'cli':321 'content':13,26 'convers':116 'creation':27 'default':324 'densiti':35,179 'depend':347,350 'descript':207,325 'detect':28 'determin':97 'etc':182 'excess':190 'extern':349 'f':340 'file':105,109,121,307,317,329 'find':149 'fl':48 'follow':87,224 'format':133,144,339,343 'found':167 'generat':12 'get':94 'high':163 'human':24,219,226,298,319 'import':197 'indic':14 'inlin':137,335 'input':99,132,304,326,328 'issu':176,185 'json':134,145,346 'length':186 'level':160 'list':168 'locat':173 'lovstudio':1 'low':161 'mandatori':84 'markdown':301 'md':331 'medium':162 'must':86 'next':205 'non':23 'non-human':22 'offer':309 'one':170 'option':206 'order':91 'output':295,296,342 'paragraph':38,180 'pass':18 'past':112 'path':106,330 'pattern':39 'phrase':32,166 'present':148 'provid':103 'python':130,140 'read':107 'refer':322 'repeat':188 'repetit':40 'request':221 'result':313 'rewrit':16,223 'risk':154,159 'rule':227 'run':127 'save':117 'score':155 'sentenc':36,184 'show':150 'signal':43 'skill':210 'skill-anti-wechat-ai-check' 'skills/lovstudio-anti-wechat-ai-check/scripts/analyze.py':131,141 'sourc':100 'source-lovstudio' 'starter':189 'stdlib':351 'step':89,92,125,146,192,217,293 'structur':175 'suffix':320 'temp':120 'templat':31,165 'text':113,124,138,142,333,336,341,344 'topic-agent-skills' 'topic-ai-coding-assistant' 'topic-cjk' 'topic-claude-code' 'topic-cursor' 'topic-gemini-cli' 'topic-markdown-to-docx' 'topic-markdown-to-pdf' 'transit':33,177 'txt':332 'uniform':37,181,187 'use':46,73,123,198 'user':102,111,152,196 'wechat':4,19,45,51 'word':34,178 'workflow':83 'write':311 '不改原文':214 '不要过度口语化':291 '专业术语保留':290 '仅查看报告':208 '你想想':280 '使用口语化表达':277 '保持原文的立场和态度':292 '保留原意':288 '偶尔使用不完整句或省略句':262 '其次':246 '具体案例':275 '减少':282 '列出每个问题的具体修改建议':213 '删掉':237 '删除不必要的':244 '删除或替换报告中标出的每个模板短语':230 '刻意制造长短句交替':252 '加入个人经历':274 '加入口语化表达':259 '去年':233 '反问句':260 '发布后':235 '句式雷同等':62 '另外':248 '基于微信公众平台运营规范':66 '增加人味':273 '字':255,258 '字使用':284 '小结':270 '并可输出人性化润色后的版本':65 '微信公众号':54 '感叹句':261 '或换成口语化的收尾':238 '打破句式均匀':251 '打破段落均匀':264 '数字细节':276 '文章内容':143 '有的段落只有一两句话':265 '有的段落可以很长':266 '条款':68 '核心观点和信息不能丢失':289 '检测文章中的':57 '模板短语':60 '此外':247 '消除模板短语':229 '生成痕迹':59 '生成的草稿改写为更自然的人类风格':82 '用具体的逻辑关系替代泛化连接词':249 '用户准备将':74 '用户想将':80 '用户想检查一篇文章是否有明显':77 '用户自己修改':209 '痕迹':79 '痕迹检测与人性化润色':56 '的':191,283 '的三段式':271 '的检测逻辑':70 '目标':241,285 '直接说具体的事':232 '直接输出修改版':215 '短句':253 '结束':211 '给出修改建议':212 '给出风险':63 '综上所述':236 '讲真':279 '论据':269 '论点':268 '评分和修改建议':64 '说白了':278 '辅助写作的文章发布到微信公众号':76 '输出人性化润色后的完整文章':216 '过渡词堆砌':61 '过渡词密度':242 '适当使用不规范但自然的表达':281 '避免每段都是':267 '长句':256 '降低过渡词密度':240 '随着科技的不断发展':231 '非真人自动化创作行为':69 '首先':245","prices":[{"id":"9b891779-ec20-4c67-ba74-bae507713bd6","listingId":"8ad01d3b-f22b-464e-b10c-e9c27acf350f","amountUsd":"0","unit":"free","nativeCurrency":null,"nativeAmount":null,"chain":null,"payTo":null,"paymentMethod":"skill-free","isPrimary":true,"details":{"org":"lovstudio","category":"skills","install_from":"skills.sh"},"createdAt":"2026-04-18T22:18:49.762Z"}],"sources":[{"listingId":"8ad01d3b-f22b-464e-b10c-e9c27acf350f","source":"github","sourceId":"lovstudio/skills/anti-wechat-ai-check","sourceUrl":"https://github.com/lovstudio/skills/tree/main/skills/anti-wechat-ai-check","isPrimary":false,"firstSeenAt":"2026-04-18T22:18:49.762Z","lastSeenAt":"2026-04-22T00:56:33.711Z"}],"details":{"listingId":"8ad01d3b-f22b-464e-b10c-e9c27acf350f","quickStartSnippet":null,"exampleRequest":null,"exampleResponse":null,"schema":null,"openapiUrl":null,"agentsTxtUrl":null,"citations":[],"useCases":[],"bestFor":[],"notFor":[],"kindDetails":{"org":"lovstudio","slug":"anti-wechat-ai-check","github":{"repo":"lovstudio/skills","stars":40,"topics":["agent-skills","ai-coding-assistant","cjk","claude-code","cursor","gemini-cli","markdown-to-docx","markdown-to-pdf"],"license":"mit","html_url":"https://github.com/lovstudio/skills","pushed_at":"2026-04-21T15:57:51Z","description":"Agent skills for AI coding assistants — Markdown to PDF/DOCX with 14 themes, CJK support","skill_md_sha":"3f2ec674676f2b587d50fc4a12a27bcd2ed75853","skill_md_path":"skills/anti-wechat-ai-check/SKILL.md","default_branch":"main","skill_tree_url":"https://github.com/lovstudio/skills/tree/main/skills/anti-wechat-ai-check"},"layout":"multi","source":"github","category":"skills","frontmatter":{"name":"lovstudio:anti-wechat-ai-check","license":"MIT","description":"Analyze articles for AI-generated content indicators and rewrite to pass WeChat's 3.27 non-human automated content creation detection. Checks for template phrases, transition word density, sentence uniformity, paragraph pattern repetition, and other signals that WeChat uses to flag AI content. Outputs a risk report and an optional humanized rewrite. Use when the user wants to check if an article looks AI-generated, make an article more human-like, bypass WeChat AI detection, or humanize AI-written content. Also trigger when the user mentions \"去AI痕迹\", \"人性化润色\", \"微信AI检测\", \"anti-ai-check\", \"humanize article\", \"公众号发文检查\".","compatibility":"Requires Python 3.8+ (stdlib only, no external dependencies). Cross-platform: macOS, Windows, Linux."},"skills_sh_url":"https://skills.sh/lovstudio/skills/anti-wechat-ai-check"},"updatedAt":"2026-04-22T00:56:33.711Z"}}