--brand {品牌名} --mode quick/home/admin/.openclaw/workspace/skills/fake-international-brand-detector/一个用于检测"假国际品牌"的智能 Agent 技能(增强版)。
通过 七项维度 验证品牌的国际化真实性(基于 NYO3 验证经验优化):
有效高质量项数量品牌类型≥5 项高质量内容✅ 真国际品牌3-4 项/混合质量⚠️ 存疑品牌≤2 项或假国际特征明显❌ 假国际品牌
"有质量的内容"标准(优化版):
验证项真国际品牌标准假国际品牌特征官网 WHOISregistrant_country=本土国家 + 多语言 + 正规备案CN/中国邮箱 (@qq.com) + 机器翻译内容销售渠道至少 2 家欧洲本土连锁药房直营/授权仅 Amazon.de/.co.uk店铺,无Boots/Holland&Barrett等销售记录≥4 个国际站点持续销售≥6 个月0-1 个站点或仅 Amazon.cn + 新注册店铺媒体报道Reuters/Forbes/WebMD/BBC报道≥5 篇无权威媒体,仅有博客/软文/中文内容WHOIS 核查registrant_country=本土国家CN/中国邮箱/地址(即使域名是.com)商标时间线海外注册时间早于或与中国同期中国市场先注册,海外后申请运营者国籍欧洲店铺由欧洲公司运营Amazon.de/.co.uk店铺由中国卖家运营
bash
# 快速检查(约 30 秒)
python /home/admin/.openclaw/workspace/skills/fake-international-brand-detector/scripts/detect_brand.py --brand "品牌名称" --mode quick
# 深度检查(约 5-10 分钟,更准确)
python /home/admin/.openclaw/workspace/skills/fake-international-brand-detector/scripts/detect_brand.py --brand "品牌名称" --mode deep
# 输出 JSON 格式(便于 API 集成)
python /home/admin/.openclaw/workspace/skills/fake-international-brand-detector/scripts/detect_brand.py --brand "NYO3" --output json
python
import requests
response = requests.post(
"http://localhost:8080/api/detect",
json={"brand": "Nike", "mode": "quick"},
headers={"Content-Type": "application/json"}
)
print(response.json())
json
{
"brand": "NYO3",
"timestamp": "2026-03-17T16:59:00+08:00",
"verifications": {
"official_website": {
"found": true,
"url": "https://nyomega.com",
"age_years": 8.5,
"languages": ["en", "de"],
"quality_score": 1.5,
"issues": [
"WHOIS registrant_country=CN",
"邮箱地址@qq.com",
"仅英文+德语,无挪威语/瑞典语"
]
},
"sales_channels": {
"found": true,
"channels": {
"physical_stores": 0,
"online_partners": [
{"platform": "amazon.de", "seller_location": "CN"},
{"platform": "amazon.co.uk", "seller_location": "CN"}
]
},
"quality_score": 2.0,
"issues": [
"无Boots UK授权",
"欧洲店铺由中国卖家运营"
]
},
"cross_border_sales": {
"found": true,
"platforms": ["amazon.com", "amazon.de"],
"order_count_estimate": ">100",
"duration_months": 3,
"quality_score": 1.5,
"issues": [
"店铺注册时间晚,销量低"
]
},
"media_exposure": {
"found": false,
"sources_checked": ["Reuters", "Forbes", "WebMD"],
"articles_found": 0,
"quality_score": 0.5,
"issues": [
"无权威媒体报道,仅有博客软文"
]
},
"whois_registration_country": {
"registrant_country": "CN",
"expected_country": "NO",
"match": false,
"email_domain": "@qq.com",
"quality_score": 0
},
"trademark_timeline": {
"china_registration_date": "201X",
"norway_registration_date": "20XX(晚于中国)",
"earlier_market": "China",
"suspicious_pattern": true,
"quality_score": 0
},
"amazon_seller_location": {
"amazon_de_seller_country": "CN",
"amazon_uk_seller_country": "CN",
"expected_countries": ["DE", "UK"],
"match": false,
"quality_score": 0
}
},
"score_summary": {
"total_items": 7,
"high_quality_count": 0,
"valid_count": 2,
"false_international_flags": [
"WHOIS显示中国注册",
"商标时间线:中国市场早于海外",
"亚马逊欧洲店铺为中国卖家运营"
]
},
"total_quality_score": 2.0,
"verdict": "🚩 假国际品牌",
"confidence": 0.95,
"recommendation": "建议避免购买或进一步调查,该品牌由中国青岛逢时科技注册和运营"
}
bash
# 运行模式
--mode quick # 快速检查(默认,约 30 秒)
--mode deep # 深度检查(约 5-10 分钟,更准确)
# 输出格式
--output json # JSON 格式
--output markdown # Markdown 报告
# API 源配置
--tavily-api-key KEY
--whois-domain tools.domaintools.com
--trademark-office CNIPA:USPTO:EUIPO
text
fake-international-brand-detector/
├── SKILL.md # 技能定义文档(此文件)
├── scripts/
│ └── detect_brand.py # 核心检测脚本(待更新)
├── reference/
│ ├── apis.md # API 和搜索引擎列表
│ ├── quality-standards.md # 内容质量标准说明
│ └── brand-examples.json # 已知品牌示例数据库(新增:NYO3案例)
└── sample/
└── test_brands.json # 测试用例数据
bash
# Tavily Web Search - 已默认配置✅
export TAVILY_API_KEY="your-key-here"
# 可选:其他 API(按需配置)
export GOOGLE_CSE_API_KEY="..."
export GOOGLE_CSE_ID="..."
WHOIS 查询:whois.domaintools.com / whois命令
商标数据库:
Wayback Machine:web.archive.org (官网历史快照)
Google News:news.google.com (国际新闻搜索)
亚马逊全球搜索:amazon.com/gp/search/
NewsAPI:newsapi.org (新闻媒体聚合)
Shopify 商店查询:shopify.store/ (电商目录)
Amazon Seller API(需登录):查看店铺运营者位置
详见 reference/apis.md 和 reference/quality-standards.md
json
{
"brand": "NOW",
"origin": "United States",
"founded": 1968,
"whois_country": "US",
"amazon_platforms": ["amazon.com", "amazon.co.uk", "amazon.de"],
"media_coverage": true,
"verdict": "✅ 真国际品牌"
}
{
"brand": "Swisse",
"origin": "Australia",
"founded": 1987,
"whois_country": "AU",
"amazon_platforms": ["amazon.com", "amazon.co.uk", "amazon.de", "amazon.fr"],
"media_coverage": true,
"verdict": "✅ 真国际品牌"
}
{
"brand": "NYO3",
"origin": "China (青岛逢时科技)",
"claimed_origin": "Norway(宣称挪威,实为假)",
"whois_country": "CN",
"trademark_timeline": {
"china_first": true,
"norway_later": true
},
"amazon_seller_location": {"de": "CN", "uk": "CN"},
"media_coverage": false,
"verdict": "🚩 假国际品牌"
}
{
"brand": "LOEON",
"origin": "China",
"founded": <5年,
"whois_country": "CN",
"amazon_platforms": ["amazon.cn"],
"media_coverage": false,
"verdict": "🚩 假国际品牌"
}
隐私合规:只公开查询品牌信息,不抓取用户数据
API 限流:遵守各平台 robots.txt 和使用条款
误判风险:小品牌可能因资料少被误判,可指定--mode deep 复测
时效性:建议每 6 个月重新验证一次结论
中文处理:对于"国货出海的假国际品牌"(如假 Nike)也能准确识别
NYO3案例学习:
判定逻辑:
日期版本更新内容2026-03-17v2.0新增 WHOIS 注册地、商标时间线、卖家国籍三项检查,基于 NYO3 案例优化2026-04-16v2.1优化大型集团品牌(如 Reckitt/MegaRed)的识别逻辑,增加集团品牌白名单验证,防止因缺乏独立官网而被误判为存疑品牌
MIT License - 自由用于商业和学术研究