intelligence-api
MCP server that wraps Shopify, Amazon, and Google Maps intelligence tools. AI agents pay autonomously in USDC on Base via x402 — no API keys or accounts needed.
README
rothy Intelligence API
An x402-powered intelligence API for Shopify, Amazon, and Google Maps data. AI agents pay per call in USDC on Base — no signup, no API keys, no human in the loop.
What This Does
| Tool | What It Returns | Price |
|---|---|---|
GET /shopify/analyze |
Products, pricing, apps, theme, collections for any Shopify store | $0.08 |
GET /shopify/products |
Paginated product catalog from any Shopify store | $0.02 |
GET /amazon/search |
Amazon search results with Opportunity Score | $0.05 |
GET /amazon/product |
Single ASIN analysis with FBA profitability estimate | $0.08 |
GET /maps/search |
Google Maps business search with Lead Quality Score | $0.05 |
GET /maps/leads |
Qualified sales leads with outreach hints | $0.10 |
Free endpoints: GET /health, GET /.well-known/mcp.json, GET /openapi.json
How x402 Payment Works
x402 is a protocol for machine-native micropayments using HTTP 402. No accounts, no rate-limit keys, no billing — agents pay per call in USDC on Base.
Flow:
Agent → GET /shopify/analyze?url=example.myshopify.com
Server → 402 Payment Required
X-PAYMENT-REQUIRED: {"price": "$0.08", "network": "base", ...}
Agent → signs USDC TransferWithAuthorization (ERC-3009)
Agent → GET /shopify/analyze?url=example.myshopify.com
X-PAYMENT: <signed-payment-header>
Server → verifies payment on Base
Server → 200 OK + intelligence data
Agents need a funded Base wallet with USDC. USDC settles directly to the server wallet — no intermediary, no escrow.
Endpoints
GET /shopify/analyze
Analyzes a Shopify store: product catalog, pricing distribution, top vendors, product types, detected apps (Klaviyo, Yotpo, ReCharge, etc.), theme detection.
Query params:
url(required) — store URL, e.g.example.myshopify.com
Example:
curl "https://api.example.com/shopify/analyze?url=gymshark.com" \
-H "X-PAYMENT: <signed-payment>"
Response:
{
"store_url": "https://gymshark.com",
"store_name": "Gymshark",
"product_count": 342,
"collection_count": 28,
"price_range": { "min": 18.00, "max": 65.00, "avg": 38.50 },
"top_vendors": [{ "vendor": "Gymshark", "count": 342 }],
"has_sale_items": true,
"sale_percentage": 23,
"detected_apps": ["Klaviyo", "Yotpo", "Sezzle"],
"detected_theme": "Dawn",
"analyzed_at": "2026-03-22T10:00:00Z"
}
GET /shopify/products
Paginated product catalog. Pages up to 250 products per call.
Query params:
url(required)page(optional, default: 1)limit(optional, default: 50, max: 250)
GET /amazon/search
Searches Amazon products and returns an Opportunity Score (0–100) for each. High score = strong demand + quality gap + good price tier.
Query params:
keyword(required)marketplace(optional:US|UK|DE|CA|AU, default:US)
Example:
curl "https://api.example.com/amazon/search?keyword=yoga+mat&marketplace=US" \
-H "X-PAYMENT: <signed-payment>"
Response:
{
"keyword": "yoga mat",
"marketplace": "US",
"total_results": 24,
"products": [
{
"asin": "B08N5WRWNW",
"title": "Premium Yoga Mat Non Slip...",
"price": 28.99,
"rating": 4.3,
"review_count": 12453,
"prime": true,
"opportunity_score": 72,
"url": "https://www.amazon.com/dp/B08N5WRWNW"
}
]
}
Opportunity Score formula:
- Reviews >1000 (proven demand): +15
- Rating <4.0 (quality gap to fill): +8–15
- Price $15–$60 (mass market sweet spot): +10
- Prime eligible: +5
GET /amazon/product
Deep analysis of a single ASIN including FBA fee/profit estimates.
Query params:
asin(required) — 10-character Amazon product IDmarketplace(optional, default:US)
Response includes FBA estimate:
{
"asin": "B08N5WRWNW",
"fba_estimate": {
"estimated_monthly_sales": 400,
"estimated_monthly_revenue": 11596,
"estimated_fba_fee": 9.84,
"estimated_profit_margin": 28,
"opportunity_tier": "medium"
}
}
GET /maps/search
Searches Google Maps businesses and scores each as a sales lead.
Query params:
query(required) — business type, e.g.coffee shopslocation(required) — city/area, e.g.Austin TXmax(optional, default: 20, max: 60)google_key(required) — your Google Places API key
Response includes Lead Quality Score:
{
"businesses": [
{
"name": "Brew & Co Coffee",
"rating": 3.8,
"website": null,
"phone": "+1-512-555-0100",
"lead_quality_score": 68,
"outreach_hints": [
"No website detected — pitch web presence / digital marketing",
"Low rating (3.8) — pitch reputation management or review improvement"
]
}
]
}
Lead Quality Score formula:
- Has website: +20
- Has phone: +10
- Rating <4.0: +8–15
- Reviews >50: +10–15
- Price level ≥2: +10
- Business OPERATIONAL: +10
GET /maps/leads
Returns only businesses above a quality score threshold, sorted by score. Best for building targeted outreach lists.
Query params:
industry(required)location(required)min_score(optional, default: 60)google_key(required)
Getting a Google Places API Key
Maps endpoints require your own Google Places API key (the server never stores it):
- Go to console.cloud.google.com
- Create or select a project
- Enable Places API
- Create an API key under APIs & Services → Credentials
- Pass it as
?google_key=YOUR_KEYon every Maps request
Agent Discovery
x402scan
Agents discover x402-enabled APIs via x402scan. The server returns proper x402 headers on every 402 response.
MCP Discovery
GET /.well-known/mcp.json
Returns the MCP plugin manifest pointing to the OpenAPI spec at /openapi.json.
OpenAI Plugin Discovery
GET /.well-known/ai-plugin.json
Compatible with ChatGPT plugin discovery format.
Running the Server
Setup
npm install
cp .env.example .env
# Edit .env: set WALLET_ADDRESS to your Base USDC wallet
Development
npm run dev
Production
npm run build
npm start
Environment Variables
WALLET_ADDRESS=0x... # Your USDC-receiving wallet on Base (required)
PORT=3000 # Default: 3000
NODE_ENV=production
x402 Payment Flow (Technical)
When an agent makes a request without payment:
HTTP/1.1 402 Payment Required
X-PAYMENT-REQUIRED: {
"version": "1",
"price": "$0.08",
"network": "base",
"payTo": "0xYourWalletAddress",
"asset": "USDC"
}
The agent:
- Signs a
TransferWithAuthorization(ERC-3009) for the USDC amount - Retries with
X-PAYMENT: <base64-encoded-signed-authorization> - Server verifies the signature on-chain and serves the response
USDC settles atomically to WALLET_ADDRESS on Base mainnet.
Error Responses
All errors return JSON:
{
"error": "Human-readable error message",
"detail": "Technical detail (optional)",
"retry_after": 30
}
| Status | Meaning |
|---|---|
| 400 | Bad request — check query params |
| 402 | Payment required — send x402 payment header |
| 503 | Upstream rate limited (Amazon) — retry after 30s |
| 502 | Upstream error (scraping failed) |
推荐服务器
Baidu Map
百度地图核心API现已全面兼容MCP协议,是国内首家兼容MCP协议的地图服务商。
Playwright MCP Server
一个模型上下文协议服务器,它使大型语言模型能够通过结构化的可访问性快照与网页进行交互,而无需视觉模型或屏幕截图。
Magic Component Platform (MCP)
一个由人工智能驱动的工具,可以从自然语言描述生成现代化的用户界面组件,并与流行的集成开发环境(IDE)集成,从而简化用户界面开发流程。
Audiense Insights MCP Server
通过模型上下文协议启用与 Audiense Insights 账户的交互,从而促进营销洞察和受众数据的提取和分析,包括人口统计信息、行为和影响者互动。
VeyraX
一个单一的 MCP 工具,连接你所有喜爱的工具:Gmail、日历以及其他 40 多个工具。
graphlit-mcp-server
模型上下文协议 (MCP) 服务器实现了 MCP 客户端与 Graphlit 服务之间的集成。 除了网络爬取之外,还可以将任何内容(从 Slack 到 Gmail 再到播客订阅源)导入到 Graphlit 项目中,然后从 MCP 客户端检索相关内容。
Kagi MCP Server
一个 MCP 服务器,集成了 Kagi 搜索功能和 Claude AI,使 Claude 能够在回答需要最新信息的问题时执行实时网络搜索。
e2b-mcp-server
使用 MCP 通过 e2b 运行代码。
Neon MCP Server
用于与 Neon 管理 API 和数据库交互的 MCP 服务器
Exa MCP Server
模型上下文协议(MCP)服务器允许像 Claude 这样的 AI 助手使用 Exa AI 搜索 API 进行网络搜索。这种设置允许 AI 模型以安全和受控的方式获取实时的网络信息。