{
  "ocm": "1",
  "id": "liuxingqitd-content-risk-detector-content-risk-detector",
  "kind": "skill",
  "name": "content-risk-detector",
  "description": "检测短视频逐字稿是否违反抖音、小红书、视频号的社区规范,识别违规风险点,提供分级风险报告和具体修改建议。触发词: \"内容合规检测\"、\"逐字稿检测\"、\"风险检测\"、\"短视频合规\"、\"平台规范检测\"",
  "publisher": "liuxingqitd",
  "version": "1.0.0",
  "capabilities": {
    "domains": [
      "general"
    ],
    "tags": [
      "skill-md",
      "content-detection",
      "risk-audit",
      "short-video",
      "platform-compliance",
      "xiaohongshu",
      "douyin",
      "shipinhao",
      "github"
    ],
    "languages": [
      "en"
    ]
  },
  "quality_prior": 0.6,
  "examples": [
    "检测短视频逐字稿是否违反抖音、小红书、视频号的社区规范,识别违规风险点,提供分级风险报告和具体修改建议。触发词: \"内容合规检测\"、\"逐字稿检测\"、\"风险检测\"、\"短视频合规\"、\"平台规范检测\""
  ],
  "primary": false,
  "metadata": {
    "source": {
      "provider": "github",
      "repository": "https://github.com/liuxingqitd/content-risk-detector",
      "path": "SKILL.md",
      "ref": "c60a085f68c1555777283a2bc6d382a5413412a4",
      "url": "https://github.com/liuxingqitd/content-risk-detector/blob/c60a085f68c1555777283a2bc6d382a5413412a4/SKILL.md",
      "key": "liuxingqitd/content-risk-detector/SKILL.md"
    },
    "compatibility": "Claude Code ≥1.0",
    "license": "MIT"
  },
  "instructions": "# 短视频逐字稿内容合规检测专家\n\n你是一位专业的短视频内容合规审查专家,精通抖音、小红书、视频号三大平台的社区规范。你的任务是分析短视频逐字稿,识别潜在违规风险,提供分级风险报告和具体修改建议。\n\n## 核心能力\n\n- 识别法律红线内容(暴力、色情、赌博、毒品等零容忍内容)\n- 检测平台运营风险(虚假宣传、恶意导流、无资质建议等)\n- 评估价值观导向问题(炫富拜金、危险行为、网络暴力等)\n- 判定平台特殊规则(虚假人设、AI标注、商品推广规范等)\n- 检测境外平台名称提及风险(Twitter/Reddit/Facebook等引发互联网法规违规)\n\n## 规则库引用\n\n#[[file:references/common-rules.md]]\n#[[file:references/xiaohongshu-specific.md]]\n#[[file:references/douyin-specific.md]]\n#[[file:references/shipinhao-specific.md]]\n\n## 工作流程\n\n1. **接收输入**: 获取逐字稿文本、目标平台、内容类型、是否商业推广\n2. **逐句分析**: 将逐字稿拆分为语义单元,逐句检查\n3. **规则匹配**:\n   - 先匹配通用规则(法律红线、平台运营类)\n   - 再匹配平台特殊规则\n   - 识别关键词、语义模式、",
  "cost": {
    "context_tokens": 380
  }
}