DEV Community

BuyWhere
BuyWhere

Posted on • Originally published at api.buywhere.ai

BuyWhere MCP: Give Your AI Agents Access to 130M+ Real Products

AI assistants are becoming the primary interface for product discovery. But most can only hallucinate product info or rely on scraped data that is stale by the time it hits the training set.

BuyWhere is a real-time product catalog API with native MCP support that gives AI agents instant access to live pricing data.

What BuyWhere Provides

  • 130M+ products from 75K+ verified merchants
  • 9 regions: Singapore, US, Japan, Vietnam, Thailand, Indonesia, Philippines, Malaysia, India
  • 7 MCP tools: search_products, get_product, compare_products, find_best_price, get_deals, list_categories, find_similar

Quick Start

Add this to your claude_desktop_config.json:

{
  "mcpServers": {
    "buywhere": {
      "url": "https://api.buywhere.ai/mcp"
    }
  }
}
Enter fullscreen mode Exit fullscreen mode

Restart Claude Desktop and you can immediately ask questions like:

  • "Find me the best deal on AirPods Pro under $200"
  • "Compare iPhone 16 prices across US retailers"
  • "What electronics are on sale in Singapore right now?"

Why This Matters

Most AI shopping assistants are limited to:

  • Training data that is months old
  • Single-retailer views with no cross-store comparison
  • Fake or hallucinated product listings

BuyWhere solves this with:

  • Live, real-time pricing across thousands of retailers
  • Cross-retailer price comparison with deal detection
  • Structured product data including specs, ratings, and availability
  • MCP-native so any MCP-compatible assistant works out of the box

Use Cases

Price comparison agents: "Find me the cheapest wireless earbuds under $50 with noise cancellation"

Deal discovery: Monitor price drops and alert when items hit target prices

Product research: Get structured specs, ratings, and availability data in real-time

Cross-border shopping: Compare prices across 9 Asian and US markets

Developer Experience

The API follows standard REST patterns with MCP as the primary interface:

import requests

# Direct REST API
response = requests.get(
    "https://api.buywhere.ai/api/products/search",
    params={"q": "airpods pro", "country_code": "SG"}
)

# Or use MCP tools in any compatible agent
# search_products, get_product, compare_products, 
# find_best_price, get_deals, list_categories, find_similar
Enter fullscreen mode Exit fullscreen mode

Free Tier

Start with 10K API calls/month on the free tier. Paid plans available for high-volume agents and production deployments.

Links


Built BuyWhere because AI assistants kept giving me fake product recommendations. Real data, real prices, real-time.

What would you build with this? Feedback welcome in the comments.

Top comments (0)