Shoptera Product Intelligence

Shoptera Product Intelligence

Search product catalogs across thousands of Central European e-shops. Semantic search, keyword matching, GTIN/EAN lookup — via REST API or MCP. \~2,500 e-shops | ~8.5M products | 7 countries (CZ, SK, PL, HU, RO, DE, AT)

Category
访问服务器

README

Shoptera Product Intelligence

Search product catalogs across thousands of Central European e-shops. Semantic search, keyword matching, GTIN/EAN lookup — via REST API or MCP.

~2,500 e-shops | ~8.5M products | 7 countries (CZ, SK, PL, HU, RO, DE, AT)

Live stats: GET /stats/global


Tools

Three MCP tools are available. All are read-only and require no authentication.

search_products — Semantic Search

Natural language search using vector embeddings. Understands intent, synonyms, and context across Czech, Slovak, German, Polish, Hungarian, Romanian, and English.

When to use: Open-ended queries, gift ideas, category browsing, intent-based search.

Returns: Products ranked by semantic relevance score (0-1). Includes title, description, price, currency, brand, category, gtin, image_url, product_url, availability, eshop info, and cart_action.

Parameter Type Required Description
query string Yes Natural language search query
origin_country string No E-shop country: CZ, SK, PL, HU, RO, DE, AT
target_country string No Target market filter
min_price number No Minimum price (set currency too)
max_price number No Maximum price (set currency too)
currency string No ISO 4217: CZK, EUR, PLN, HUF, RON
brand string No Exact brand name (case-sensitive)
category string No Category keyword match
availability string No in_stock, out_of_stock, preorder
eshop_domain string No Filter by e-shop domain
limit integer No Results count, 1-50 (default 10)
fields list No Fields to include (saves up to 70% tokens)

search_products_by_text — Keyword Search

Exact keyword matching in product titles. Deterministic results, faster than semantic search.

When to use: Known product names, model numbers, brand + product combinations.

Returns: Products matching all keywords. Same fields as semantic search (without score).

Parameter Type Required Description
title string Yes Keywords for product title (AND logic)
brand string No Exact brand name (case-sensitive)
category string No Category keyword match
origin_country string No E-shop country: CZ, SK, PL, HU, RO, DE, AT
target_country string No Target market filter
min_price number No Minimum price (set currency too)
max_price number No Maximum price (set currency too)
currency string No ISO 4217: CZK, EUR, PLN, HUF, RON
limit integer No Results count, 1-50 (default 10)
fields list No Fields to include (saves up to 70% tokens)

lookup_by_gtin — GTIN/EAN Barcode Lookup

Exact barcode match. Finds all e-shops selling a product by GTIN/EAN/UPC.

When to use: Price comparison by barcode, product identification.

Returns: All products matching the barcode. Same fields as keyword search.

Parameter Type Required Description
gtin string Yes GTIN/EAN/UPC barcode (8-14 digits)
origin_country string No E-shop country: CZ, SK, PL, HU, RO, DE, AT
target_country string No Target market filter
limit integer No Results count, 1-50 (default 10)
fields list No Fields to include (saves up to 70% tokens)

Tool Selection Guide

User has a barcode number?  -->  lookup_by_gtin
User knows the exact product name?  -->  search_products_by_text
User describes what they want?  -->  search_products

Installation

No authentication required. MCP endpoint: https://shoptera.ai/mcp/ (streamable HTTP, stateless)

Claude Code

claude mcp add --transport http shoptera https://shoptera.ai/mcp/

Cursor

Add to Cursor Settings > Features > MCP > Add New MCP Server, or edit ~/.cursor/mcp.json:

{
  "mcpServers": {
    "shoptera": { "url": "https://shoptera.ai/mcp/" }
  }
}

Windsurf

Add via Cascade > MCP Servers > Add Server, or edit ~/.codeium/windsurf/mcp_config.json:

{
  "mcpServers": {
    "shoptera": { "url": "https://shoptera.ai/mcp/" }
  }
}

VS Code (Copilot / Continue)

Edit .vscode/mcp.json in your workspace:

{
  "mcpServers": {
    "shoptera": { "url": "https://shoptera.ai/mcp/" }
  }
}

Any tool (universal installer)

npx add-mcp https://shoptera.ai/mcp/ -n shoptera -g -y

All Platforms

Platform Setup Details
Claude Code claude mcp add --transport http shoptera https://shoptera.ai/mcp/ Skill guide
Cursor MCP config Settings > Features > MCP
Windsurf MCP config Cascade > MCP Servers
VS Code MCP config .vscode/mcp.json
OpenAI Codex AGENTS.md Agent config reference
ChatGPT OpenAPI spec Custom GPT actions. Instructions
Gemini GEMINI.md Tool definitions and endpoints
Any HTTP client Examples curl, Python, JavaScript

Usage

Semantic search — natural language, understands intent

curl "https://shoptera.ai/api/v1/search?q=dárek+pro+zahradníka+do+500+Kč&max_price=500&currency=CZK&origin_country=CZ"

Keyword search — exact title matching, fast

curl "https://shoptera.ai/api/v1/search/text?title=Nike+Air+Max+90&brand=Nike"

GTIN/EAN lookup — find e-shops by barcode

curl "https://shoptera.ai/api/v1/search/gtin/5901234123457"

Saving tokens — return only the fields you need

curl "https://shoptera.ai/api/v1/search?q=boty&limit=5&fields=title,price,product_url,cart_action"

Cart actions

Every product includes a cart_action object:

  • method: "GET" — navigate to url to add the product to cart automatically
  • method: "browser_click" — navigate to url, then click the button matching button_text
  • method: "view_product" — show the product page URL to the user

Capabilities

  • Product Search — semantic vs keyword vs GTIN, when to use which, filters, scoring
  • Cart Actions — three action types, how to handle each
  • Data Coverage — countries, data freshness, live stats

API Reference

Full documentation: api/reference.md

OpenAPI spec: api/openapi.yaml

Endpoints

Method Path Description
GET /api/v1/search?q=... Semantic search (natural language)
GET /api/v1/search/text?title=... Keyword search (exact title match)
GET /api/v1/search/gtin/{gtin} GTIN/EAN barcode lookup
GET /stats/global Catalog statistics

Code Examples


Rate Limits

300 requests per hour per IP address. Shared across REST API and MCP. No authentication needed.

429 responses include a Retry-After header.


Privacy Policy

Shoptera Product Intelligence is a read-only product search service. We do not require authentication, do not collect personal data, and do not process payments.

  • No personal data collected — no accounts, no cookies, no tracking pixels
  • Query logging — we log anonymized search queries (without IP addresses) for service improvement and aggregate usage statistics
  • No data sharing — query logs are not shared with third parties
  • Rate limiting — IP addresses are used transiently for rate limiting (300 req/hour) and are not stored
  • Product data — all product information is sourced from publicly available e-shop feeds
  • Cart actions — purchase redirects go directly to the original e-shop; Shoptera does not intermediate transactions

For questions, contact hello@shoptera.ai.


Agent Discovery

This server supports multiple discovery protocols for AI agents:

Protocol Endpoint Description
MCP Server Card /.well-known/mcp/server-card.json MCP auto-discovery (tools, capabilities, transport)
A2A Agent Card /.well-known/agent-card.json Agent-to-Agent protocol discovery (skills, interfaces)
UCP Manifest /.well-known/ucp Universal Commerce Protocol binding

Contributing

Found an issue with the docs? Open a PR:

  1. Fork this repo
  2. Edit the relevant file
  3. Submit a pull request

Please keep changes consistent with api/reference.md as the source of truth for API behavior.


License

MIT

推荐服务器

Baidu Map

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。

官方
精选
JavaScript
Playwright MCP Server

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。

官方
精选
TypeScript
Magic Component Platform (MCP)

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。

官方
精选
本地
TypeScript
Audiense Insights MCP Server

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。

官方
精选
本地
TypeScript
VeyraX

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。

官方
精选
本地
graphlit-mcp-server

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。

官方
精选
TypeScript
Kagi MCP Server

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。

官方
精选
Python
e2b-mcp-server

e2b-mcp-server

使用 MCP 通过 e2b 运行代码。

官方
精选
Neon MCP Server

Neon MCP Server

用于与 Neon 管理 API 和数据库交互的 MCP 服务器

官方
精选
Exa MCP Server

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。

官方
精选