POST https://scrapeapi.pangolinfo.com/api/v1/scrape

Headers:
  Content-Type: application/json
  Authorization: Bearer {YOUR_API_KEY}

Body（亚马逊商品页示例）:
{
  "url": "https://www.amazon.com/dp/{ASIN}",
  "parserName": "amazonProduct",   // 使用亚马逊商品专用解析器
  "country": "us"                  // 指定亚马逊站点
}

import requests
import json
from typing import Optional


class PangolinAmazonTool:
    """
    OpenClaw 数据工具：Pangolinfo Amazon Scraper API 封装
    为 Agent 提供实时亚马逊商品数据
    """

    BASE_URL = "https://scrapeapi.pangolinfo.com/api/v1/scrape"

    def __init__(self, api_key: str):
        self.headers = {
            "Content-Type": "application/json",
            "Authorization": f"Bearer {api_key}"
        }

    def get_product(self, asin: str, country: str = "us") -> dict:
        """
        获取单个 ASIN 的实时商品数据

        Args:
            asin: 亚马逊商品编号（如 B09G9FPHY6）
            country: 站点代码（us / uk / de / jp 等）

        Returns:
            包含价格、BSR、评分、Buy Box 等字段的结构化数据
        """
        url = f"https://www.amazon.{country}/dp/{asin}"
        payload = {
            "url": url,
            "parserName": "amazonProduct",
            "country": country
        }

        resp = requests.post(
            self.BASE_URL,
            headers=self.headers,
            json=payload,
            timeout=30
        )
        resp.raise_for_status()
        result = resp.json()

        if result.get("code") != 0:
            raise ValueError(f"API 错误: {result.get('message')}")

        return self._extract_key_fields(result["data"])

    def compare_competitors(
        self,
        asins: list[str],
        country: str = "us"
    ) -> list[dict]:
        """
        批量对比竞品数据

        Args:
            asins: ASIN 列表（建议 ≤10 个）
            country: 站点代码

        Returns:
            按 BSR 排序的竞品数据列表
        """
        results = []
        for asin in asins:
            try:
                data = self.get_product(asin, country)
                data["asin"] = asin
                results.append(data)
            except Exception as e:
                results.append({"asin": asin, "error": str(e)})

        # 按 BSR 升序排列（排名越小越靠前）
        return sorted(
            [r for r in results if "bsr" in r],
            key=lambda x: x.get("bsr", 9999999)
        )

    def _extract_key_fields(self, data: dict) -> dict:
        """提取 Agent 最常使用的核心字段"""
        return {
            "title": data.get("title", ""),
            "price": data.get("price", {}).get("current"),
            "rating": data.get("rating"),
            "review_count": data.get("reviewCount"),
            "bsr": data.get("bestSellersRank", [{}])[0].get("rank"),
            "bsr_category": data.get("bestSellersRank", [{}])[0].get("category"),
            "buy_box_winner": data.get("buyBox", {}).get("seller"),
            "in_stock": data.get("availability") == "In Stock",
            "seller_count": data.get("offerCount"),
            "scraped_at": data.get("scrapedAt")
        }


# ==================== OpenClaw Tool 注册示例 ====================

# 初始化工具
amazon_tool = PangolinAmazonTool(api_key="YOUR_PANGOLINFO_API_KEY")

# 示例：让 OpenClaw Agent 调用竞品对比
competitor_asins = ["B09G9FPHY6", "B0BTSR8T9M", "B0CKQSQ2WS"]
report = amazon_tool.compare_competitors(competitor_asins, country="us")

# OpenClaw Agent 接收到的上下文（JSON 格式，LLM 可直接分析）
for item in report:
    print(f"ASIN: {item['asin']}")
    print(f"  价格: ${item.get('price', 'N/A')}")
    print(f"  BSR:  #{item.get('bsr', 'N/A')} in {item.get('bsr_category', '')}")
    print(f"  评分: {item.get('rating')} ({item.get('review_count')} reviews)")
    print(f"  Buy Box: {item.get('buy_box_winner', '未知')}")
    print(f"  数据时间: {item.get('scraped_at')}")
    print()