应用场景

为 Agent 工作流而生

从自主研究到多 Agent 协作 — 为你电商体系中的每个 Agent 提供动力。

差别在哪？

人类：每天 100 个产品。Agent + ZooData：每天 10,000+ 个产品。

自主选品 Agent

让 Agent 持续扫描市场，发现蓝海机会。人类每天能看 100 个产品。Agent 能看 10,000 个。

人类：每天 100 个产品。Agent：每天 10,000 个产品。

自动化市场机会扫描
14 种内置选品预设
多维度筛选逻辑

查看 API 文档

agent.py

from apiclaw import APIClawClient

client = APIClawClient(api_key="hms_live_xxx")

# Discover blue ocean opportunities
products = client.products.search(
    category_path=["Electronics", "Headphones"],
    mode="blue_ocean",  # 14 preset modes available
    filters={
        "monthly_revenue_min": 10000,
        "review_count_max": 100,
        "rating_min": 4.0
    }
)

# Your agent processes 10,000+ results
for product in products:
    analyze_opportunity(product)

竞争情报网络

多 Agent 协同监控竞争对手。价格变动、新品上架、评论变化 — 全部秒级感知。

7x24 监控，价格变动秒级通知。

实时价格变动通知
新品上架检测
评论趋势监控

查看 API 文档

agent.py

# Multi-dimensional competitor lookup
competitors = client.competitors.lookup(
    keyword="wireless earbuds",
    brand="Sony",
    asin="B09V3KXJPB"  # Any dimension
)

# Real-time monitoring
for competitor in competitors:
    product = client.realtime.product(
        asin=competitor["asin"]
    )

    if price_changed(product):
        alert_agent(product)

动态定价 Agent

基于实时市场信号自动调价。这不是定价 SaaS — 而是你定价 Agent 的数据层。

不是定价 SaaS，是定价 Agent 的数据层。

实时竞品价格追踪
价格带分析
利润优化信号

查看 API 文档

agent.py

# Batch real-time price lookup
asins = ["B09V3KXJPB", "B08N5WRWNW", ...]

for asin in asins:
    product = client.realtime.product(asin=asin)

    market_price = product["price"]
    competitor_prices = get_competitor_prices(asin)

    # Your agent decides optimal price
    optimal_price = pricing_agent.decide(
        current_price=my_price,
        market_price=market_price,
        competitor_prices=competitor_prices
    )

大规模评论智能

从海量评论数据中提取产品洞察。百万条评论变成 10 个可执行洞察。Token 成本降低 95%。

百万条评论 → 10 个洞察。Token 成本降 95%。

大规模情感分析
功能需求提取
客户痛点识别

查看 API 文档

agent.py

# AI-powered review analysis
insights = client.reviews.analyze(
    asins=["B09V3KXJPB"],
    analysis_types=[
        "sentiment",
        "keywords",
        "pain_points",
        "feature_requests"
    ]
)

# Pre-processed, structured output
# No need to run your own NLP
print(insights["top_complaints"])
print(insights["feature_gaps"])
print(insights["sentiment_score"])

市场雷达

持续监控品类趋势，捕捉上升信号。流动的信号，不是静态的报告。

流动的信号，不是静态的报告。

品类趋势追踪
BSR 波动预警
新兴机会发现

查看 API 文档

agent.py

# Market trend monitoring
def daily_scan():
    categories = get_watch_list()

    for category in categories:
        market = client.markets.search(
            category_path=category,
            sample_type="by_sale_100"
        )

        # BSR updated every 15 minutes
        signals = extract_signals(market)

        if signals["trending_up"]:
            notify_agent(category, signals)

# Run continuously
schedule.every(15).minutes.do(daily_scan)

端到端 Listing 自动化

从发现到上架的全自动流程。OpenClaw 生态：数据层 + 执行层。

OpenClaw 生态：数据层 + 执行层。

自动化产品发现
Listing 内容生成
库存同步自动化

查看 API 文档

agent.py

# End-to-end automation workflow
class ListingAgent:
    def run(self, category):
        # Step 1: Discover opportunities
        products = client.products.search(
            category_path=category,
            mode="emerging"
        )

        # Step 2: Validate with real-time data
        for product in products:
            live = client.realtime.product(
                asin=product["asin"]
            )

            if self.validate(live):
                # Step 3: Generate listing
                listing = self.generate_listing(live)

                # Step 4: Publish (via OpenClaw)
                self.publish(listing)

与你的 AI 技术栈无缝集成

OpenAPI 3.0 规范意味着一键导入任何框架

LangChain

CrewAI

AutoGen

Claude MCP

OpenAI Functions

Custom Agents

开始构建探索 Skills

from apiclaw import APIClawClient client = APIClawClient(api_key="hms_live_xxx") # Discover blue ocean opportunities products = client.products.search( category_path=["Electronics", "Headphones"], mode="blue_ocean", # 14 preset modes available filters={ "monthly_revenue_min": 10000, "review_count_max": 100, "rating_min": 4.0 } ) # Your agent processes 10,000+ results for product in products: analyze_opportunity(product)

# Multi-dimensional competitor lookup competitors = client.competitors.lookup( keyword="wireless earbuds", brand="Sony", asin="B09V3KXJPB" # Any dimension ) # Real-time monitoring for competitor in competitors: product = client.realtime.product( asin=competitor["asin"] ) if price_changed(product): alert_agent(product)

# Batch real-time price lookup asins = ["B09V3KXJPB", "B08N5WRWNW", ...] for asin in asins: product = client.realtime.product(asin=asin) market_price = product["price"] competitor_prices = get_competitor_prices(asin) # Your agent decides optimal price optimal_price = pricing_agent.decide( current_price=my_price, market_price=market_price, competitor_prices=competitor_prices )

# AI-powered review analysis insights = client.reviews.analyze( asins=["B09V3KXJPB"], analysis_types=[ "sentiment", "keywords", "pain_points", "feature_requests" ] ) # Pre-processed, structured output # No need to run your own NLP print(insights["top_complaints"]) print(insights["feature_gaps"]) print(insights["sentiment_score"])

# Market trend monitoring def daily_scan(): categories = get_watch_list() for category in categories: market = client.markets.search( category_path=category, sample_type="by_sale_100" ) # BSR updated every 15 minutes signals = extract_signals(market) if signals["trending_up"]: notify_agent(category, signals) # Run continuously schedule.every(15).minutes.do(daily_scan)

# End-to-end automation workflow class ListingAgent: def run(self, category): # Step 1: Discover opportunities products = client.products.search( category_path=category, mode="emerging" ) # Step 2: Validate with real-time data for product in products: live = client.realtime.product( asin=product["asin"] ) if self.validate(live): # Step 3: Generate listing listing = self.generate_listing(live) # Step 4: Publish (via OpenClaw) self.publish(listing)