{
  "skill_name": "topic-selector",
  "description": "AI互动剧选题筛选 - 数据采集配置",
  "version": "1.0",
  "created": "2026-03-27",

  "data_sources": [
    {
      "name": "WETRUE短剧数据",
      "url": "https://wetrue.com",
      "type": "充值榜单",
      "priority": "P0",
      "update_frequency": "每日",
      "crawl_config": {
        "method": "GET",
        "headers": {
          "User-Agent": "Mozilla/5.0 (Macintosh; Intel Mac OS X 10_15_7) AppleWebKit/537.36"
        },
        "interval": 3000,
        "timeout": 15000
      },
      "parsing_rules": {
        "target_selector": ".drama-list, .ranking-list",
        "fields": ["rank", "title", "charge_amount", "genre", "platform"]
      },
      "robots_allowed": true
    },
    {
      "name": "鞭牛士短剧新闻",
      "url": "https://www.bianews.com/news/duanju",
      "type": "行业新闻",
      "priority": "P0",
      "update_frequency": "实时",
      "crawl_config": {
        "method": "GET",
        "headers": {
          "User-Agent": "Mozilla/5.0 (Macintosh; Intel Mac OS X 10_15_7) AppleWebKit/537.36"
        },
        "interval": 3000,
        "timeout": 15000
      },
      "parsing_rules": {
        "target_selector": ".news-list, article",
        "fields": ["title", "summary", "url", "publish_time"]
      },
      "robots_allowed": true
    },
    {
      "name": "抖音热榜",
      "url": "https://www.douyin.com/aweme/v1/web/hot/search/",
      "type": "平台趋势",
      "priority": "P1",
      "update_frequency": "每小时",
      "crawl_config": {
        "method": "GET",
        "headers": {
          "User-Agent": "Mozilla/5.0 (Macintosh; Intel Mac OS X 10_15_7) AppleWebKit/537.36"
        },
        "interval": 5000,
        "timeout": 15000
      },
      "parsing_rules": {
        "target_selector": "word",
        "fields": ["rank", "word", "hot_value"]
      },
      "robots_allowed": true
    },
    {
      "name": "快手热榜",
      "url": "https://www.kuaishou.com/short-video/lists/hot",
      "type": "平台趋势",
      "priority": "P1",
      "update_frequency": "每小时",
      "crawl_config": {
        "method": "GET",
        "headers": {
          "User-Agent": "Mozilla/5.0 (Macintosh; Intel Mac OS X 10_15_7) AppleWebKit/537.36"
        },
        "interval": 5000,
        "timeout": 15000
      },
      "parsing_rules": {
        "target_selector": ".hot-list",
        "fields": ["rank", "title", "views"]
      },
      "robots_allowed": true
    },
    {
      "name": "36kr短剧资讯",
      "url": "https://36kr.com/search/news/keyword/%E7%9F%AD%E5%89%A7",
      "type": "科技媒体",
      "priority": "P1",
      "update_frequency": "每日",
      "crawl_config": {
        "method": "GET",
        "headers": {
          "User-Agent": "Mozilla/5.0 (Macintosh; Intel Mac OS X 10_15_7) AppleWebKit/537.36"
        },
        "interval": 3000,
        "timeout": 15000
      },
      "parsing_rules": {
        "target_selector": ".search-result, .article-list",
        "fields": ["title", "summary", "url", "publish_time"]
      },
      "robots_allowed": true
    }
  ],

  "genre_classification": {
    "战神赘婿": ["复仇", "逆袭", "身份反转", "窝囊女婿"],
    "甜宠": ["恋爱", "总裁", "豪门", "甜妻"],
    "虐恋": ["复仇", "误会", "虐心", "生死"],
    "都市": ["职场", "家庭", "情感", "成长"],
    "玄幻": ["修仙", "系统", "穿越", "异界"],
    "悬疑": ["推理", "惊悚", "侦破", "烧脑"]
  },

  "quality_metrics": {
    "完播率": {"excellent": ">40%", "good": ">30%", "fair": ">20%"},
    "付费率": {"excellent": ">5%", "good": ">3%", "fair": ">1%"},
    "互动率": {"excellent": ">3%", "good": ">2%", "fair": ">1%"},
    "次留率": {"excellent": ">30%", "good": ">25%", "fair": ">20%"}
  },

  "crawl_schedule": {
    "daily": [
      {"time": "09:00", "sources": ["WETRUE", "鞭牛士"], "action": "榜单+新闻"},
      {"time": "15:00", "sources": ["WETRUE"], "action": "榜单更新"},
      {"time": "21:00", "sources": ["WETRUE", "鞭牛士"], "action": "榜单+新闻"}
    ],
    "weekly": [
      {"day": "周一", "action": "周度趋势分析报告"}
    ]
  },

  "rate_limiting": {
    "min_interval_ms": 2000,
    "max_requests_per_minute": 30,
    "retry_on_failure": true,
    "max_retries": 3
  },

  "data_retention": {
    "raw_data_days": 90,
    "aggregated_data_days": 365
  }
}