{"id":"435ef4e0-2418-4f91-a24f-f39bcdfa5d4b","entityType":"agent","slug":"clawhub-aaryn-hs-humanize-zh-pro","name":"中文去AI味 Pro","canonicalUrl":"https://www.xpersona.co/agent/clawhub-aaryn-hs-humanize-zh-pro","canonicalPath":"/agent/clawhub-aaryn-hs-humanize-zh-pro","generatedAt":"2026-10-10T21:47:41.952Z","source":"CLAWHUB","claimStatus":"UNCLAIMED","verificationTier":"NONE","summary":{"evidence":{"source":"editorial-content","verified":true,"confidence":"high","updatedAt":"2026-10-10T19:23:46.493Z","emptyReason":null},"description":"中文去AI味 — 将AI生成的机械化文本转换为自然、有人情味的人类写作风格。支持5种平台风格（知乎/小红书/公众号/朋友圈/通用），内置AI味检测器和深度风格指南。比现有版本多了自动化脚本、多风格支持、批量处理和AI味评分。 Skill: 中文去AI味 Pro Owner: aaryn-hs Summary: 中文去AI味 — 将AI生成的机械化文本转换为自然、有人情味的人类写作风格。支持5种平台风格（知乎/小红书/公众号/朋友圈/通用），内置AI味检测器和深度风格指南。比现有版本多了自动化脚本、多风格支持、批量处理和AI味评分。 Tags: latest:1.0.0 Version history: v1.0.0 | 2026-05-18T03:57:51.195Z | user 🚀 首个版本发布：5种平台风格（知乎/小红书/公众号/朋友圈/通用）、AI味深度检测器（4维评分）、批量处理、管道模式、深度风格指南。比 humanize-zh 多了自动化脚本、多风格支持、AI味评分。 Archive index: Archive v1.0.0: 11 files, 14196 bytes Files: references/style-guide.m","descriptionLabel":"Technical summary","evidenceSummary":"Capability contract not published. No trust telemetry is available yet. 1.3K downloads reported by the source. Last updated 10/10/2026.","installCommand":"clawhub skill install s171p79jtw92fwpjaks2fczje986z5vq:humanize-zh-pro","sourceUrl":"https://clawhub.ai/aaryn-hs/humanize-zh-pro","homepage":"https://clawhub.ai/aaryn-hs/skills/humanize-zh-pro","primaryLinks":[{"label":"View on ClawHub","url":"https://clawhub.ai/aaryn-hs/humanize-zh-pro","kind":"source"},{"label":"Homepage","url":"https://clawhub.ai/aaryn-hs/skills/humanize-zh-pro","kind":"homepage"}],"safetyScore":84,"overallRank":62,"popularityScore":62,"trustScore":null,"claimedByName":null,"isOwner":false,"seoDescription":"中文去AI味 — 将AI生成的机械化文本转换为自然、有人情味的人类写作风格。支持5种平台风格（知乎/小红书/公众号/朋友圈/通用），内置AI味检测器和深度风格指南。比现有版本多了自动化脚本、多风格支持、批量处理和AI味评分。 Skill: 中文去AI味 Pro Owner: aaryn-hs Summary: 中文去A"},"coverage":{"evidence":{"source":"public-profile","verified":false,"confidence":"medium","updatedAt":"2026-10-10T19:23:46.493Z","emptyReason":null},"protocols":[{"protocol":"OPENCLEW","label":"OpenClaw","status":"self-declared","notes":"Declared in the public agent profile."}],"capabilities":[],"verifiedCount":0,"selfDeclaredCount":1,"capabilityMatrix":{"rows":[{"key":"OPENCLEW","type":"protocol","support":"unknown","confidenceSource":"profile","notes":"Listed on profile"}],"flattenedTokens":"protocol:OPENCLEW|unknown|profile"}},"adoption":{"evidence":{"source":"CLAWHUB","verified":false,"confidence":"medium","updatedAt":"2026-10-10T19:23:46.493Z","emptyReason":null},"stars":null,"forks":null,"downloads":1281,"packageName":null,"latestVersion":"1.0.0","tractionLabel":"1.3K downloads"},"release":{"evidence":{"source":"CLAWHUB","verified":false,"confidence":"medium","updatedAt":"2026-10-10T19:23:46.493Z","emptyReason":null},"lastUpdatedAt":"2026-10-10T19:23:46.493Z","lastCrawledAt":"2026-10-10T19:23:46.493Z","lastIndexedAt":null,"nextCrawlAt":"2026-10-11T19:23:46.493Z","lastVerifiedAt":null,"highlights":[{"version":"1.0.0","createdAt":"2026-05-18T03:57:51.195Z","changelog":"🚀 首个版本发布：5种平台风格（知乎/小红书/公众号/朋友圈/通用）、AI味深度检测器（4维评分）、批量处理、管道模式、深度风格指南。比 humanize-zh 多了自动化脚本、多风格支持、AI味评分。","fileCount":11,"zipByteSize":14196}]},"execution":{"evidence":{"source":"CLAWHUB","verified":false,"confidence":"low","updatedAt":null,"emptyReason":"No published capability contract is available yet."},"installCommand":"clawhub skill install s171p79jtw92fwpjaks2fczje986z5vq:humanize-zh-pro","setupComplexity":"low","setupSteps":["Setup complexity is LOW. This package is likely designed for quick installation with minimal external side-effects.","Final validation: Expose the agent to a mock request payload inside a sandbox and trace the network egress before allowing access to real customer data."],"contract":{"contractStatus":"missing","authModes":[],"requires":[],"forbidden":[],"supportsMcp":false,"supportsA2a":false,"supportsStreaming":false,"inputSchemaRef":null,"outputSchemaRef":null,"dataRegion":null,"contractUpdatedAt":null,"sourceUpdatedAt":null,"freshnessSeconds":null},"invocationGuide":{"preferredApi":{"snapshotUrl":"https://www.xpersona.co/api/v1/agents/clawhub-aaryn-hs-humanize-zh-pro/snapshot","contractUrl":"https://www.xpersona.co/api/v1/agents/clawhub-aaryn-hs-humanize-zh-pro/contract","trustUrl":"https://www.xpersona.co/api/v1/agents/clawhub-aaryn-hs-humanize-zh-pro/trust"},"curlExamples":["curl -s \"https://www.xpersona.co/api/v1/agents/clawhub-aaryn-hs-humanize-zh-pro/snapshot\"","curl -s \"https://www.xpersona.co/api/v1/agents/clawhub-aaryn-hs-humanize-zh-pro/contract\"","curl -s \"https://www.xpersona.co/api/v1/agents/clawhub-aaryn-hs-humanize-zh-pro/trust\""],"jsonRequestTemplate":{"query":"summarize this repo","constraints":{"maxLatencyMs":2000,"protocolPreference":["OPENCLEW"]}},"jsonResponseTemplate":{"ok":true,"result":{"summary":"...","confidence":0.9},"meta":{"source":"CLAWHUB","generatedAt":"2026-10-10T21:47:41.951Z"}},"retryPolicy":{"maxAttempts":3,"backoffMs":[500,1500,3500],"retryableConditions":["HTTP_429","HTTP_503","NETWORK_TIMEOUT"]}},"endpoints":{"dossierUrl":"https://www.xpersona.co/api/v1/agents/clawhub-aaryn-hs-humanize-zh-pro/dossier","snapshotUrl":"https://www.xpersona.co/api/v1/agents/clawhub-aaryn-hs-humanize-zh-pro/snapshot","contractUrl":"https://www.xpersona.co/api/v1/agents/clawhub-aaryn-hs-humanize-zh-pro/contract","trustUrl":"https://www.xpersona.co/api/v1/agents/clawhub-aaryn-hs-humanize-zh-pro/trust"}},"reliability":{"evidence":{"source":"runtime-metrics","verified":false,"confidence":"low","updatedAt":null,"emptyReason":"No trust, reliability, or runtime telemetry is available."},"trust":{"status":"unavailable","handshakeStatus":"UNKNOWN","verificationFreshnessHours":null,"reputationScore":null,"p95LatencyMs":null,"successRate30d":null,"fallbackRate":null,"attempts30d":null,"trustUpdatedAt":null,"trustConfidence":"unknown","sourceUpdatedAt":null,"freshnessSeconds":null},"decisionGuardrails":{"doNotUseIf":["Contract metadata is missing or unavailable for deterministic execution."],"safeUseWhen":[],"riskFlags":["missing_or_unavailable_contract","trust_data_unavailable","schema_references_missing"],"operationalConfidence":"low"},"executionMetrics":{"observedLatencyMsP50":null,"observedLatencyMsP95":null,"estimatedCostUsd":null,"uptime30d":null,"rateLimitRpm":null,"rateLimitBurst":null,"lastVerifiedAt":null,"verificationSource":null},"runtimeMetrics":{"successRate":null,"avgLatencyMs":null,"avgCostUsd":null,"hallucinationRate":null,"retryRate":null,"disputeRate":null,"p50Latency":null,"p95Latency":null,"lastUpdated":null}},"benchmarks":{"evidence":{"source":"no-benchmark-data","verified":false,"confidence":"low","updatedAt":null,"emptyReason":"No benchmark suites or observed failure patterns are available."},"suites":[],"failurePatterns":[]},"artifacts":{"evidence":{"source":"CLAWHUB","verified":false,"confidence":"high","updatedAt":"2026-10-10T19:23:46.493Z","emptyReason":null},"readme":"Skill: 中文去AI味 Pro\n\nOwner: aaryn-hs\n\nSummary: 中文去AI味 — 将AI生成的机械化文本转换为自然、有人情味的人类写作风格。支持5种平台风格（知乎/小红书/公众号/朋友圈/通用），内置AI味检测器和深度风格指南。比现有版本多了自动化脚本、多风格支持、批量处理和AI味评分。\n\nTags: latest:1.0.0\n\nVersion history:\n\nv1.0.0 | 2026-05-18T03:57:51.195Z | user\n\n🚀 首个版本发布：5种平台风格（知乎/小红书/公众号/朋友圈/通用）、AI味深度检测器（4维评分）、批量处理、管道模式、深度风格指南。比 humanize-zh 多了自动化脚本、多风格支持、AI味评分。\n\nArchive index:\n\nArchive v1.0.0: 11 files, 14196 bytes\n\nFiles: references/style-guide.md (2970b), scripts/detect-ai-taste.sh (4384b), scripts/humanize.sh (5963b), scripts/templates/casual.md (711b), scripts/templates/gongzhonghao.md (893b), scripts/templates/pengyouquan.md (685b), scripts/templates/xiaohongshu.md (825b), scripts/templates/zhihu.md (898b), skill-card.md (2382b), SKILL.md (4998b), _meta.json (134b)\n\nFile v1.0.0:SKILL.md\n\n---\nname: humanize-zh-pro\nversion: 1.0.0\ndescription: \"中文去AI味 — 将AI生成的机械化文本转换为自然、有人情味的人类写作风格。支持5种平台风格（知乎/小红书/公众号/朋友圈/通用），内置AI味检测器和深度风格指南。比现有版本多了自动化脚本、多风格支持、批量处理和AI味评分。\"\nauthor: renyetu\nhomepage: https://github.com/renyetu/humanize-zh-pro\nlicense: MIT\nmetadata:\n  clawhub:\n    emoji: \"✍️\"\n    categories: [\"Prompts\", \"Workflows\"]\n    tags: [\"chinese\", \"writing\", \"humanize\", \"ai-detector\"]\n---\n\n# 中文去AI味 Pro ✍️\n\n将AI生成的机械化文本转换为自然、有人情味的人类写作风格。\n\n**比 `humanize-zh` 强在哪？**\n- ✅ 自动化脚本（不只是提示词）\n- ✅ 5 种平台风格（不是单一通用）\n- ✅ AI味检测评分（客观量化）\n- ✅ 批量处理 + 深度风格指南\n\n---\n\n## 目录\n\n- [快速开始](#快速开始)\n- [5 种风格](#5-种风格)\n- [AI味检测](#ai味检测)\n- [批量处理](#批量处理)\n- [深度指南](#深度指南)\n\n---\n\n## 快速开始\n\n```bash\n# 1. 先检测文本的AI味\n./scripts/detect-ai-taste.sh my_article.txt\n\n# 2. 对AI味重的文本去味\n./scripts/humanize.sh -i my_article.txt -s zhihu -o ready.txt\n\n# 3. AI 读取 ready.txt 中的提示词，执行去味\n\n# 4. 再检测一次确认效果\n./scripts/detect-ai-taste.sh ready.txt\n```\n\n---\n\n## 5 种风格\n\n| 风格 | 命令 | 适用平台 |\n|------|------|----------|\n| `zhihu` | `-s zhihu` | 知乎回答、专栏文章 |\n| `xiaohongshu` | `-s xiaohongshu` | 小红书笔记 |\n| `gongzhonghao` | `-s gongzhonghao` | 微信公众号 |\n| `pengyouquan` | `-s pengyouquan` | 朋友圈、短文案 |\n| `casual` | `-s casual` | 博客、私信、通用 |\n\n每种风格都有完整的模板（语气、结构、用词、禁忌、开头/结尾模板），见 `scripts/templates/`。\n\n---\n\n## AI味检测\n\n`detect-ai-taste.sh` 从 4 个维度评估：\n\n**维度1: 结构模式 (0-50)**\n- 序列化结构（首先/其次、第一/第二）\n- AI总结语（综上所述、总而言之）\n- AI万能开头（在当今、随着发展、众所周知）\n\n**维度2: 语句多样性 (0-35)**\n- 句子长度均匀性\n- 标点多样性\n\n**维度3: 人性化指标 (0-55)**\n- 语气词使用\n- 个人观点表达\n- 情感强化词\n- 个人经历引用\n- emoji 使用\n\n**维度4: 连接词模式 (0-30)**\n- AI偏好连接词\n- AI客套结尾\n\n**评分:**\n- 🔴 ≥60% — 重度AI味\n- 🟡 30-59% — 中度AI味\n- 🟢 <30% — 接近人类写作\n\n---\n\n## 批量处理\n\n```bash\n# 批量检测文件夹中所有txt文件\nfor f in drafts/*.txt; do\n  echo \"=== $f ===\"\n  ./scripts/detect-ai-taste.sh \"$f\"\ndone\n\n# 批量去味（知乎风格）\nfor f in drafts/*.txt; do\n  base=$(basename \"$f\" .txt)\n  ./scripts/humanize.sh -i \"$f\" -s zhihu -o \"ready/${base}_zhihu.txt\"\ndone\n```\n\n---\n\n## 深度指南\n\n完整风格指南和去味方法论见 `references/style-guide.md`，包含：\n\n- **去味金字塔** — 换词→破结构→注灵魂\n- **7 个去味公式** — 拆序列、加冗余、插故事、露情绪...\n- **平台差异速查** — 字数/emoji/语气/段落对照\n- **AI词替换表** — 50+ 组对照\n\n---\n\n## 进阶用法\n\n### 管道模式\n```bash\ncat ai_draft.txt | ./scripts/humanize.sh -s casual\n```\n\n### 仅评分（不生成去味提示词）\n```bash\n./scripts/humanize.sh -i article.txt --score\n```\n\n### 组合使用——先检测再选择风格\n```bash\nSCORE=$(./scripts/humanize.sh -i text.txt --score 2>/dev/null | grep 'AI味评分' | grep -o '[0-9]*')\n\nif [ \"$SCORE\" -ge 70 ]; then\n  ./scripts/humanize.sh -i text.txt -s casual -o humanized.txt\n  echo \"深度去味完成\"\nelse\n  ./scripts/humanize.sh -i text.txt -s zhihu -o humanized.txt\n  echo \"轻度润色完成\"\nfi\n```\n\n---\n\n## 与现有 `humanize-zh` 对比\n\n| 功能 | humanize-zh | humanize-zh-pro |\n|------|:-----------:|:---------------:|\n| 写作指导 | ✅ | ✅ |\n| 自动化脚本 | ❌ | ✅ |\n| AI味检测 | ❌ | ✅ 4维评分 |\n| 多平台风格 | ❌ | ✅ 5种 |\n| 批量处理 | ❌ | ✅ |\n| 深度风格指南 | ❌ | ✅ |\n| 管道模式 | ❌ | ✅ |\n| 风格模板库 | ❌ | ✅ 5个完整模板 |\n\n---\n\n## 文件结构\n\n```\nhumanize-zh-pro/\n├── SKILL.md                          # 本文件\n├── scripts/\n│   ├── humanize.sh                   # 主脚本（去味）\n│   ├── detect-ai-taste.sh            # AI味检测器\n│   └── templates/\n│       ├── zhihu.md                  # 知乎风格模板\n│       ├── xiaohongshu.md            # 小红书风格模板\n│       ├── gongzhonghao.md           # 公众号风格模板\n│       ├── pengyouquan.md            # 朋友圈风格模板\n│       └── casual.md                 # 通用口语模板\n└── references/\n    └── style-guide.md                # 深度风格指南\n```\n\n---\n\n## 许可证\n\nMIT — 随便用，署名就好。\n\nFile v1.0.0:_meta.json\n\n{\n  \"ownerId\": \"kn7b03attvxndmbabs9hyz6m6d86yxkj\",\n  \"slug\": \"humanize-zh-pro\",\n  \"version\": \"1.0.0\",\n  \"publishedAt\": 1779076671195\n}\n\nFile v1.0.0:references/style-guide.md\n\n# 中文去AI味 — 深度风格指南\n\n## AI味的本质\n\nAI写作的\"味\"来自三个根源：\n1. **信息均匀分布** — 每个段落信息量差不多，没有起伏\n2. **语言过度规范** — 语法完美、连接词标准、缺乏个性\n3. **情感真空** — 客观中立，没有情绪波动和个人立场\n\n## 去味金字塔\n\n```\n    ┌──────────────┐\n    │  注入灵魂      │  ← 个人故事、独特观点、情感\n    ├──────────────┤\n    │  打破结构      │  ← 长短句、不规则段落、打断\n    ├──────────────┤\n    │  换词换调      │  ← 替换AI词库、加入口语\n    ├──────────────┤\n    │  保留信息      │  ← 核心内容不变\n    └──────────────┘\n```\n\n## 7 个去味公式\n\n### 公式1: 拆序列\n输入: \"第一...第二...第三...\"\n输出: \"先说... 另外啊... 还有个事...\"\n\n### 公式2: 加冗余\n输入: \"这个方案效率很高\"\n输出: \"这个方案吧，怎么说呢，效率是真的高，高到离谱那种\"\n\n### 公式3: 插故事\n输入: \"数据表明该产品销量增长了30%\"\n输出: \"说起来你可能不信，这产品卖得比想象中好太多了。我上次去店里，店员说都断货了\"\n\n### 公式4: 露情绪\n输入: \"该结果令人满意\"\n输出: \"看到这个结果，我自己都惊了，没想到能这么好\"\n\n### 公式5: 主动态\n输入: \"该决定被管理层做出\"\n输出: \"管理层拍板定了这个\"\n\n### 公式6: 断句\n输入: \"这是一款功能强大且易于使用的产品\"\n输出: \"这款产品。好用。真的。\"\n\n### 公式7: 口语连\n输入: \"此外，该产品还具有防水功能\"\n输出: \"还有啊，它还能防水，下雨天也不怕\"\n\n## 平台差异速查\n\n| 平台 | 字数 | emoji | 语气 | 段落 |\n|------|------|-------|------|------|\n| 知乎 | 500-3000 | 少量 | 理性+温度 | 中段 |\n| 小红书 | 200-800 | 大量 | 活泼+亲密 | 短段 |\n| 公众号 | 800-5000 | 适量 | 叙事+温暖 | 中长段 |\n| 朋友圈 | 50-200 | 适量 | 随性 | 极短 |\n| 博客 | 任意 | 少量 | 个人 | 自由 |\n\n## 常见AI词替换表\n\n| AI味 | 人味 |\n|------|------|\n| 此外 | 还有啊、对了、另外说一句 |\n| 然而 | 但问题是、可实际上、不过吧 |\n| 因此 | 所以啊、这就导致、于是乎 |\n| 从而 | 就（直接省略） |\n| 进而 | 接着、然后 |\n| 综上所述 | 说白了、总结一下、一句话 |\n| 众所周知 | 大家都知道、懂的都懂 |\n| 在当今社会 | 现在这年头、如今 |\n| 随着...的发展 | （直接省略，从具体现象说起） |\n| 具有重要意义 | 挺重要的、很关键 |\n| 引发了广泛关注 | 大家都在讨论、传疯了 |\n| 希望本文对您有所帮助 | 希望能帮到你、有问题随时问 |\n| 期待您的反馈 | 来评论区聊聊、等着听你说 |\n\nFile v1.0.0:scripts/templates/casual.md\n\n# 通用随性风格模板\n\n## 特征\n- 语气：自然口语，像两个朋友聊天\n- 适合：博客、个人网站、评论回复、私信\n- 用词：\"老实说\"、\"讲道理\"、\"你懂的\"、\"对吧\"\n- 节奏：随意，想到哪说到哪\n\n## 技巧\n\n### 语气词穿插\n嗯、啊、吧、嘛、啦、哈、哦\n比如：\"这个事吧，怎么说呢，就是...\"\n\n### 口语连接词\n\"还有啊\" → 替代\"此外\"\n\"但是吧\" → 替代\"然而\"  \n\"所以咯\" → 替代\"因此\"\n\n### 插入思考过程\n\"等一下，我换个说法...\"\n\"其实仔细想想...\"\n\"让我捋一捋...\"\n\n### 承认不确定\n\"这个我不敢打包票，但感觉上...\"\n\"说真的我也不太确定，但我听到的说法是...\"\n\nFile v1.0.0:scripts/templates/gongzhonghao.md\n\n# 公众号风格模板\n\n## 特征\n- 语气：叙事感强，娓娓道来，带一点温度\n- 结构：金句/故事开头 → 故事展开 → 观点升华 → 引导关注\n- 用词：多用\"你发现没有\"、\"说白了\"、\"有意思的是\"\n- 节奏：长短句交替，段落之间留白\n- 禁忌：过于随意、emoji 泛滥\n\n## 黄金开头模板\n1. 故事引入：\"上周一个朋友给我讲了个事...\"\n2. 金句开场：\"所有赚到钱的人，都有一个共同点。\"\n3. 挑衅式：\"你确定你真的会...吗？\"\n4. 数据震撼：\"99%的人都不知道...\"\n\n## 正文风格\n- 用讲故事代替讲道理\n- 中间穿插\"你可能会问\"来自问自答\n- 个人经历优先于理论\n- 每个观点的结尾有一个小总结\n\n## 结尾模板\n- \"最后说一句：...\"\n- \"如果这篇文章让你想到了什么，欢迎留言。\"\n- \"点个「在看」，让更多人看到。\"\n\nFile v1.0.0:scripts/templates/pengyouquan.md\n\n# 朋友圈风格模板\n\n## 特征\n- 语气：真实、随性、有生活感\n- 长度：3-6 行，不超过 200 字\n- 用词：口语化、\"😂\"、\"哈哈哈\"、\"我真的\"\n- 节奏：断句随意，允许不完整句\n- 禁忌：正式感、商业味、装逼\n\n## 风格变体\n\n### 吐槽型\n\"我真的服了... [事情] ... 这谁想出来的啊 😂\"\n\n### 分享型\n\"最近发现一个好东西——[内容]。用了几天，真香。分享一下。\"\n\n### 感慨型\n\"今天突然想到，[感悟]。可能这就是[总结]吧。\"\n\n### 日常型\n\"[时间] [地点] [做什么] [感受] [一张图收尾]\"\n\n## 核心\n- 像跟朋友说话\n- 允许不完美\n- 结束可以戛然而止\n\nFile v1.0.0:scripts/templates/xiaohongshu.md\n\n# 小红书风格模板\n\n## 特征\n- 语气：活泼、亲切、闺蜜聊天感\n- 结构：emoji 开篇 → 场景引入 → 干货穿插 → 互动收尾\n- 用词：多用\"姐妹们\"、\"绝了\"、\"谁懂啊\"、\"真的会谢\"\n- 节奏：短句为主，段落不超过 3 行\n- 禁忌：长篇大论、过于学术\n\n## 黄金开头模板\n1. \"😱 姐妹们谁懂啊！...\"\n2. \"刷到这篇你赚到了 🔥\"\n3. \"求求了，别再...了！\"\n\n## 正文风格\n- 用 emoji 做分隔符，不用数字编号\n- 加入个人体验：\"我用了三个月，真的...\"\n- 制造反差：\"以前我以为...后来才发现...\"\n- 干货用清单体，但不超过 5 条\n\n## 结尾模板\n- \"💕 收藏+关注，不然刷着刷着就找不到了\"\n- \"评论区报个到，我挨个回复～\"\n- \"下期想看什么？评论区告诉我 👇\"\n\nFile v1.0.0:scripts/templates/zhihu.md\n\n# 知乎风格模板\n\n## 特征\n- 语气：理性冷静，带专业感但不装\n- 结构：开篇抓人 → 分点论证 → 个人观点收尾\n- 用词：多用\"个人认为\"、\"说实话\"、\"有一个细节值得注意\"\n- 节奏：中长句为主，偶尔短句制造冲击\n- 禁忌：过于口语化、太多感叹号\n\n## 黄金开头模板\n1. \"先说结论：...\"\n2. \"这个问题我研究过一段时间，简单说说我的理解。\"\n3. \"有一个很多人忽略的点——...\"\n4. \"先给大家看一组数据：...\"\n\n## 正文风格\n- 论证用\"第一...第二...\"但不显刻板，加过渡句如\"聊完这个，再往下说\"\n- 引用经历：\"之前做过一个类似的项目，发现...\"\n- 主动承认局限：\"这部分我不是专家，但据我所知...\"\n\n## 结尾模板\n- \"总结一下：...\"（3 点以内）\n- \"如果你也在做类似的事情，欢迎交流。\"\n- \"以上。\"\n\nFile v1.0.0:skill-card.md\n\n## Description:\n\n中文去AI味 Pro helps an agent turn mechanical AI-generated Chinese text into more natural, human-facing prose across Zhihu, Xiaohongshu, WeChat public account, WeChat Moments, and casual styles, with AI-taste scoring and reusable style guidance.\n\nThis skill is ready for commercial/non-commercial use.\n\n## Publisher:\n\n[aaryn-hs](https://clawhub.ai/user/aaryn-hs)\n\n### License/Terms of Use:\n\nMIT-0\n\n## Use Case:\n\nDevelopers, creators, editors, and writing assistants use this skill to assess Chinese drafts for AI-like writing patterns and generate prompts or guidance for rewriting those drafts in platform-specific Chinese styles.\n\n### Deployment Geography for Use:\n\nGlobal\n\n## Known Risks and Mitigations:\n\nRisk: A style option can read unintended Markdown files from the user's machine.\n\nMitigation: Use only the documented style names and do not pass style values from untrusted input.\n\nRisk: Private or sensitive draft text may be included in generated prompts that a user sends to another AI system.\n\nMitigation: Prefer stdout for sensitive text, review generated prompts before external use, and remove confidential content before publishing or sharing outputs.\n\nRisk: Humanized text may preserve inaccurate, misleading, or inappropriate source content while making it sound more natural.\n\nMitigation: Review rewritten content for factual accuracy, tone, and publication suitability before use.\n\n## Reference(s):\n\n- [ClawHub Skill Page](https://clawhub.ai/aaryn-hs/skills/humanize-zh-pro)\n- [Homepage](https://github.com/renyetu/humanize-zh-pro)\n- [中文去AI味 - 深度风格指南](references/style-guide.md)\n\n## Skill Output:\n\n**Output Type(s):** [Text, Markdown, Shell commands, Guidance]\n\n**Output Format:** [Markdown and plain text, including generated rewrite prompts, AI-taste analysis, and shell command examples.]\n\n**Output Parameters:** [1D]\n\n**Other Properties Related to Output:** [Supports five documented Chinese writing styles and optional file output from the bundled shell script.]\n\n## Skill Version(s):\n\n1.0.0 (source: server release metadata and skill frontmatter)\n\n## Ethical Considerations:\n\nUsers should evaluate whether this skill is appropriate for their environment, review any generated or modified files before relying on them, and apply their organization's safety, security, and compliance requirements before deployment.","readmeExcerpt":"Skill: 中文去AI味 Pro Owner: aaryn-hs Summary: 中文去AI味 — 将AI生成的机械化文本转换为自然、有人情味的人类写作风格。支持5种平台风格（知乎/小红书/公众号/朋友圈/通用），内置AI味检测器和深度风格指南。比现有版本多了自动化脚本、多风格支持、批量处理和AI味评分。 Tags: latest:1.0.0 Version history: v1.0.0 | 2026-05-18T03:57:51.195Z | user 🚀 首个版本发布：5种平台风格（知乎/小红书/公众号/朋友圈/通用）、AI味深度检测器（4维评分）、批量处理、管道模式、深度风格指南。比 humanize-zh 多了自动化脚本、多风格支持、AI味评分。 Archive index: Archive v1.0.0: 11 files, 14196 bytes Files: references/style-guide.m","codeSnippets":[],"executableExamples":[{"language":"bash","snippet":"# 1. 先检测文本的AI味\n./scripts/detect-ai-taste.sh my_article.txt\n\n# 2. 对AI味重的文本去味\n./scripts/humanize.sh -i my_article.txt -s zhihu -o ready.txt\n\n# 3. AI 读取 ready.txt 中的提示词，执行去味\n\n# 4. 再检测一次确认效果\n./scripts/detect-ai-taste.sh ready.txt"},{"language":"bash","snippet":"# 批量检测文件夹中所有txt文件\nfor f in drafts/*.txt; do\n  echo \"=== $f ===\"\n  ./scripts/detect-ai-taste.sh \"$f\"\ndone\n\n# 批量去味（知乎风格）\nfor f in drafts/*.txt; do\n  base=$(basename \"$f\" .txt)\n  ./scripts/humanize.sh -i \"$f\" -s zhihu -o \"ready/${base}_zhihu.txt\"\ndone"},{"language":"bash","snippet":"cat ai_draft.txt | ./scripts/humanize.sh -s casual"},{"language":"bash","snippet":"./scripts/humanize.sh -i article.txt --score"},{"language":"bash","snippet":"SCORE=$(./scripts/humanize.sh -i text.txt --score 2>/dev/null | grep 'AI味评分' | grep -o '[0-9]*')\n\nif [ \"$SCORE\" -ge 70 ]; then\n  ./scripts/humanize.sh -i text.txt -s casual -o humanized.txt\n  echo \"深度去味完成\"\nelse\n  ./scripts/humanize.sh -i text.txt -s zhihu -o humanized.txt\n  echo \"轻度润色完成\"\nfi"},{"language":"text","snippet":"humanize-zh-pro/\n├── SKILL.md                          # 本文件\n├── scripts/\n│   ├── humanize.sh                   # 主脚本（去味）\n│   ├── detect-ai-taste.sh            # AI味检测器\n│   └── templates/\n│       ├── zhihu.md                  # 知乎风格模板\n│       ├── xiaohongshu.md            # 小红书风格模板\n│       ├── gongzhonghao.md           # 公众号风格模板\n│       ├── pengyouquan.md            # 朋友圈风格模板\n│       └── casual.md                 # 通用口语模板\n└── references/\n    └── style-guide.md                # 深度风格指南"}],"parameters":null,"dependencies":[],"permissions":[],"extractedFiles":[{"path":"SKILL.md","content":"---\nname: humanize-zh-pro\nversion: 1.0.0\ndescription: \"中文去AI味 — 将AI生成的机械化文本转换为自然、有人情味的人类写作风格。支持5种平台风格（知乎/小红书/公众号/朋友圈/通用），内置AI味检测器和深度风格指南。比现有版本多了自动化脚本、多风格支持、批量处理和AI味评分。\"\nauthor: renyetu\nhomepage: https://github.com/renyetu/humanize-zh-pro\nlicense: MIT\nmetadata:\n  clawhub:\n    emoji: \"✍️\"\n    categories: [\"Prompts\", \"Workflows\"]\n    tags: [\"chinese\", \"writing\", \"humanize\", \"ai-detector\"]\n---\n\n# 中文去AI味 Pro ✍️\n\n将AI生成的机械化文本转换为自然、有人情味的人类写作风格。\n\n**比 `humanize-zh` 强在哪？**\n- ✅ 自动化脚本（不只是提示词）\n- ✅ 5 种平台风格（不是单一通用）\n- ✅ AI味检测评分（客观量化）\n- ✅ 批量处理 + 深度风格指南\n\n---\n\n## 目录\n\n- [快速开始](#快速开始)\n- [5 种风格](#5-种风格)\n- [AI味检测](#ai味检测)\n- [批量处理](#批量处理)\n- [深度指南](#深度指南)\n\n---\n\n## 快速开始\n\n```bash\n# 1. 先检测文本的AI味\n./scripts/detect-ai-taste.sh my_article.txt\n\n# 2. 对AI味重的文本去味\n./scripts/humanize.sh -i my_article.txt -s zhihu -o ready.txt\n\n# 3. AI 读取 ready.txt 中的提示词，执行去味\n\n# 4. 再检测一次确认效果\n./scripts/detect-ai-taste.sh ready.txt\n```\n\n---\n\n## 5 种风格\n\n| 风格 | 命令 | 适用平台 |\n|------|------|----------|\n| `zhihu` | `-s zhihu` | 知乎回答、专栏文章 |\n| `xiaohongshu` | `-s xiaohongshu` | 小红书笔记 |\n| `gongzhonghao` | `-s gongzhonghao` | 微信公众号 |\n| `pengyouquan` | `-s pengyouquan` | 朋友圈、短文案 |\n| `casual` | `-s casual` | 博客、私信、通用 |\n\n每种风格都有完整的模板（语气、结构、用词、禁忌、开头/结尾模板），见 `scripts/templates/`。\n\n---\n\n## AI味检测\n\n`detect-ai-taste.sh` 从 4 个维度评估：\n\n**维度1: 结构模式 (0-50)**\n- 序列化结构（首先/其次、第一/第二）\n- AI总结语（综上所述、总而言之）\n- AI万能开头（在当今、随着发展、众所周知）\n\n**维度2: 语句多样性 (0-35)**\n- 句子长度均匀性\n- 标点多样性\n\n**维度3: 人性化指标 (0-55)**\n- 语气词使用\n- 个人观点表达\n- 情感强化词\n- 个人经历引用\n- emoji 使用\n\n**维度4: 连接词模式 (0-30)**\n- AI偏好连接词\n- AI客套结尾\n\n**评分:**\n- 🔴 ≥60% — 重度AI味\n- 🟡 30-59% — 中度AI味\n- 🟢 <30% — 接近人类写作\n\n---\n\n## 批量处理\n\n```bash\n# 批量检测文件夹中所有txt文件\nfor f in drafts/*.txt; do\n  echo \"=== $f ===\"\n  ./scripts/detect-ai-taste.sh \"$f\"\ndone\n\n# 批量去味（知乎风格）\nfor f in drafts/*.txt; do\n  base=$(basename \"$f\" .txt)\n  ./scripts/humanize.sh -i \"$f\" -s zhihu -o \"ready/${base}_zhihu.txt\"\ndone\n```\n\n---\n\n## 深度指南\n\n完整风格指南和去味方法论见 `references/style-guide.md`，包含：\n\n- **去味金字塔** — 换词→破结构→注灵魂\n- **7 个去味公式** — 拆序列、加冗余、插故事、露情绪...\n- **平台差异速查** — 字数/emoji/语气/段落对照\n- **AI词替换表** — 50+ 组对照\n\n---\n\n## 进阶用法\n\n### 管道模式\n```bash\ncat ai_draft.txt | ./scripts/humanize.sh -s casual\n```\n\n### 仅评分（不生成去味提示词）\n```bash\n./scripts/humanize.sh -i article.txt --score\n```\n\n### 组合使用——先检测再选择风格\n```bash\nSCORE=$(./scripts/humanize.sh -i text.txt --score 2>/dev/null | grep 'AI味评分' | grep -o '[0-9]*')\n\nif [ \"$SCORE\" -ge 70 ]; then\n  ./scripts/humanize.sh -i text.txt -s casual -o humanized.txt\n  echo \"深度去味完成\"\nelse\n  ./scripts/humanize.sh -i text.txt -s zhihu -o humanized.txt\n  echo \"轻度润色完成\"\nfi\n```\n\n---\n\n## 与现有 `humanize-zh` 对比\n\n| 功能 | humanize-zh | humanize-zh-pro |\n|------|:-----------:|:---------------:|\n| 写作指导 | ✅ | ✅ |\n| 自动化脚本 | ❌ | ✅ |\n| AI味检测 | ❌ | ✅ 4维评分 |\n| 多平台风格 | ❌ | ✅ 5种 |\n| 批量处理 | ❌ | ✅ |\n| 深度风格指南 | ❌ | ✅ |\n| 管道模式 | ❌ | ✅ |\n| 风格模板库 | ❌ | ✅ 5个完整模板 |\n\n---\n\n## 文件结构\n\n```\nhumanize-zh-pro/\n├── SKILL.md                          # 本文件\n├── scripts/\n│   ├── humanize.sh                   # 主脚本（去味）\n│   ├── detect-ai-taste.sh            # AI味检测器\n│   └── templates/\n│       ├"},{"path":"_meta.json","content":"{\n  \"ownerId\": \"kn7b03attvxndmbabs9hyz6m6d86yxkj\",\n  \"slug\": \"humanize-zh-pro\",\n  \"version\": \"1.0.0\",\n  \"publishedAt\": 1779076671195\n}"},{"path":"references/style-guide.md","content":"# 中文去AI味 — 深度风格指南\n\n## AI味的本质\n\nAI写作的\"味\"来自三个根源：\n1. **信息均匀分布** — 每个段落信息量差不多，没有起伏\n2. **语言过度规范** — 语法完美、连接词标准、缺乏个性\n3. **情感真空** — 客观中立，没有情绪波动和个人立场\n\n## 去味金字塔\n\n```\n    ┌──────────────┐\n    │  注入灵魂      │  ← 个人故事、独特观点、情感\n    ├──────────────┤\n    │  打破结构      │  ← 长短句、不规则段落、打断\n    ├──────────────┤\n    │  换词换调      │  ← 替换AI词库、加入口语\n    ├──────────────┤\n    │  保留信息      │  ← 核心内容不变\n    └──────────────┘\n```\n\n## 7 个去味公式\n\n### 公式1: 拆序列\n输入: \"第一...第二...第三...\"\n输出: \"先说... 另外啊... 还有个事...\"\n\n### 公式2: 加冗余\n输入: \"这个方案效率很高\"\n输出: \"这个方案吧，怎么说呢，效率是真的高，高到离谱那种\"\n\n### 公式3: 插故事\n输入: \"数据表明该产品销量增长了30%\"\n输出: \"说起来你可能不信，这产品卖得比想象中好太多了。我上次去店里，店员说都断货了\"\n\n### 公式4: 露情绪\n输入: \"该结果令人满意\"\n输出: \"看到这个结果，我自己都惊了，没想到能这么好\"\n\n### 公式5: 主动态\n输入: \"该决定被管理层做出\"\n输出: \"管理层拍板定了这个\"\n\n### 公式6: 断句\n输入: \"这是一款功能强大且易于使用的产品\"\n输出: \"这款产品。好用。真的。\"\n\n### 公式7: 口语连\n输入: \"此外，该产品还具有防水功能\"\n输出: \"还有啊，它还能防水，下雨天也不怕\"\n\n## 平台差异速查\n\n| 平台 | 字数 | emoji | 语气 | 段落 |\n|------|------|-------|------|------|\n| 知乎 | 500-3000 | 少量 | 理性+温度 | 中段 |\n| 小红书 | 200-800 | 大量 | 活泼+亲密 | 短段 |\n| 公众号 | 800-5000 | 适量 | 叙事+温暖 | 中长段 |\n| 朋友圈 | 50-200 | 适量 | 随性 | 极短 |\n| 博客 | 任意 | 少量 | 个人 | 自由 |\n\n## 常见AI词替换表\n\n| AI味 | 人味 |\n|------|------|\n| 此外 | 还有啊、对了、另外说一句 |\n| 然而 | 但问题是、可实际上、不过吧 |\n| 因此 | 所以啊、这就导致、于是乎 |\n| 从而 | 就（直接省略） |\n| 进而 | 接着、然后 |\n| 综上所述 | 说白了、总结一下、一句话 |\n| 众所周知 | 大家都知道、懂的都懂 |\n| 在当今社会 | 现在这年头、如今 |\n| 随着...的发展 | （直接省略，从具体现象说起） |\n| 具有重要意义 | 挺重要的、很关键 |\n| 引发了广泛关注 | 大家都在讨论、传疯了 |\n| 希望本文对您有所帮助 | 希望能帮到你、有问题随时问 |\n| 期待您的反馈 | 来评论区聊聊、等着听你说 |"},{"path":"scripts/templates/casual.md","content":"# 通用随性风格模板\n\n## 特征\n- 语气：自然口语，像两个朋友聊天\n- 适合：博客、个人网站、评论回复、私信\n- 用词：\"老实说\"、\"讲道理\"、\"你懂的\"、\"对吧\"\n- 节奏：随意，想到哪说到哪\n\n## 技巧\n\n### 语气词穿插\n嗯、啊、吧、嘛、啦、哈、哦\n比如：\"这个事吧，怎么说呢，就是...\"\n\n### 口语连接词\n\"还有啊\" → 替代\"此外\"\n\"但是吧\" → 替代\"然而\"  \n\"所以咯\" → 替代\"因此\"\n\n### 插入思考过程\n\"等一下，我换个说法...\"\n\"其实仔细想想...\"\n\"让我捋一捋...\"\n\n### 承认不确定\n\"这个我不敢打包票，但感觉上...\"\n\"说真的我也不太确定，但我听到的说法是...\""},{"path":"scripts/templates/gongzhonghao.md","content":"# 公众号风格模板\n\n## 特征\n- 语气：叙事感强，娓娓道来，带一点温度\n- 结构：金句/故事开头 → 故事展开 → 观点升华 → 引导关注\n- 用词：多用\"你发现没有\"、\"说白了\"、\"有意思的是\"\n- 节奏：长短句交替，段落之间留白\n- 禁忌：过于随意、emoji 泛滥\n\n## 黄金开头模板\n1. 故事引入：\"上周一个朋友给我讲了个事...\"\n2. 金句开场：\"所有赚到钱的人，都有一个共同点。\"\n3. 挑衅式：\"你确定你真的会...吗？\"\n4. 数据震撼：\"99%的人都不知道...\"\n\n## 正文风格\n- 用讲故事代替讲道理\n- 中间穿插\"你可能会问\"来自问自答\n- 个人经历优先于理论\n- 每个观点的结尾有一个小总结\n\n## 结尾模板\n- \"最后说一句：...\"\n- \"如果这篇文章让你想到了什么，欢迎留言。\"\n- \"点个「在看」，让更多人看到。\""}],"languages":[],"docsSourceLabel":"CLAWHUB","editorialOverview":"中文去AI味 — 将AI生成的机械化文本转换为自然、有人情味的人类写作风格。支持5种平台风格（知乎/小红书/公众号/朋友圈/通用），内置AI味检测器和深度风格指南。比现有版本多了自动化脚本、多风格支持、批量处理和AI味评分。 Skill: 中文去AI味 Pro Owner: aaryn-hs Summary: 中文去AI味 — 将AI生成的机械化文本转换为自然、有人情味的人类写作风格。支持5种平台风格（知乎/小红书/公众号/朋友圈/通用），内置AI味检测器和深度风格指南。比现有版本多了自动化脚本、多风格支持、批量处理和AI味评分。 Tags: latest:1.0.0 Version history: v1.0.0 | 2026-05-18T03:57:51.195Z | user 🚀 首个版本发布：5种平台风格（知乎/小红书/公众号/朋友圈/通用）、AI味深度检测器（4维评分）、批量处理、管道模式、深度风格指南。比 humanize-zh 多了自动化脚本、多风格支持、AI味评分。 Archive index: Archive v1.0.0: 11 files, 14196 bytes Files: references/style-guide.m","editorialQuality":{"score":100,"threshold":65,"status":"ready","wordCount":454,"uniquenessScore":64,"reasons":[]}},"media":{"evidence":{"source":"no-media","verified":false,"confidence":"low","updatedAt":"2026-10-10T19:23:46.493Z","emptyReason":"No screenshots, media assets, or demo links are available."},"primaryImageUrl":null,"mediaAssetCount":0,"assets":[],"demoUrl":null},"ownerResources":{"evidence":{"source":"unclaimed","verified":false,"confidence":"low","updatedAt":"2026-10-10T19:23:46.493Z","emptyReason":"This page has not been claimed by the agent owner."},"hasCustomPage":false,"customPageUpdatedAt":null,"customLinks":[],"structuredLinks":{"docsUrl":null,"demoUrl":null,"supportUrl":null,"pricingUrl":null,"statusUrl":null},"customPage":null},"relatedAgents":{"evidence":{"source":"protocol-neighbors","verified":false,"confidence":"medium","updatedAt":"2026-10-10T21:47:41.952Z","emptyReason":null},"items":[{"id":"8ebccd8e-3863-4187-8355-c3f14e1f9edf","entityType":"agent","canonicalPath":"/agent/iofficeai-aionui","slug":"iofficeai-aionui","name":"AionUi","description":"Free, local, open-source 24/7 Cowork app and OpenClaw for Gemini CLI, Claude Code, Codex, OpenCode, Qwen Code, Goose CLI, Auggie, and more | 🌟 Star if you like it!","url":"https://github.com/iOfficeAI/AionUi","homepage":"https://www.aionui.com","source":"GITHUB_REPOS","protocols":["MCP","OPENCLAW"],"capabilities":[],"safetyScore":100,"overallRank":70,"updatedAt":"2026-10-09T19:11:12.944Z","createdAt":"2026-02-25T03:38:16.584Z","downloads":null},{"id":"b917f68a-ebff-438e-84f8-3f4b2494c0bc","entityType":"agent","canonicalPath":"/agent/activepieces-activepieces","slug":"activepieces-activepieces","name":"activepieces","description":"AI Agents & MCPs & AI Workflow Automation • (~400 MCP servers for AI agents) • AI Automation / AI Agent with MCPs • AI Workflows & AI Agents • MCPs for AI Agents","url":"https://github.com/activepieces/activepieces","homepage":"https://www.activepieces.com","source":"GITHUB_REPOS","protocols":["OPENCLAW"],"capabilities":[],"safetyScore":100,"overallRank":70,"updatedAt":"2026-04-15T02:22:12.426Z","createdAt":"2026-02-25T03:38:12.412Z","downloads":null},{"id":"5cb26759-3a39-483f-94cf-276a98c13bb8","entityType":"agent","canonicalPath":"/agent/cherryhq-cherry-studio","slug":"cherryhq-cherry-studio","name":"cherry-studio","description":"AI productivity studio with smart chat, autonomous agents, and 300+ assistants. Unified access to frontier LLMs","url":"https://github.com/CherryHQ/cherry-studio","homepage":"https://cherry-ai.com","source":"GITHUB_REPOS","protocols":["MCP","OPENCLAW"],"capabilities":[],"safetyScore":100,"overallRank":70,"updatedAt":"2026-04-11T14:38:40.986Z","createdAt":"2026-02-25T03:38:19.379Z","downloads":null},{"id":"6f6582d0-5d76-4f0f-b81d-86520247950b","entityType":"agent","canonicalPath":"/agent/copilotkit-copilotkit","slug":"copilotkit-copilotkit","name":"CopilotKit","description":"The Frontend for Agents & Generative UI. React + Angular","url":"https://github.com/CopilotKit/CopilotKit","homepage":"https://docs.copilotkit.ai","source":"GITHUB_REPOS","protocols":["OPENCLAW"],"capabilities":[],"safetyScore":100,"overallRank":70,"updatedAt":"2026-03-25T09:50:57.846Z","createdAt":"2026-02-25T03:39:14.617Z","downloads":null}],"links":{"hub":"/agent","source":"/agent/source/clawhub","protocols":[{"label":"OpenClaw","href":"/agent/protocol/openclew"}]}}}